On August 12, 2008, about 1017 eastern daylight time, a Beech G35, N4615D, registered to a private individual, operated by the pilot as "Angel Flight 15 Delta" experienced an in-flight loss of control while maneuvering to re-intercept a final approach course, and then crashed into a parking lot of a shopping center located in South Easton, Massachusetts. Instrument meteorological conditions prevailed at the time and an instrument flight rules (IFR) flight plan was filed for the 14 Code of Federal Regulations (CFR) Part 91 personal flight that departed Francis S. Gabreski Airport (FOK), Westhampton, New York, about 0909, destined for General Edward Lawrence Logan International Airport (BOS), Boston, Massachusetts. The airplane was destroyed, and the certificated commercial pilot and the two passengers were killed.

The pilot was flying the passengers to BOS so one of the passengers could receive medical treatment in the Boston area.

After takeoff the pilot established contact with New York Terminal Radar Approach Control, and then with Providence Air Traffic Control Tower, while proceeding towards the destination airport. Air traffic control (ATC) communications were then transferred to Boston Consolidated Terminal Radar Approach Control (Boston Consolidated TRACON).

According to a transcription of communications with Boston Consolidated TRACON Initial Departure Position, the pilot contacted the facility at 0954:05, and advised the controller that he had automated terminal information service (ATIS) information Papa. The controller instructed the pilot to expect the instrument landing system (ILS) approach to runway 4R, and provided the altimeter setting. The flight continued towards the destination airport and the controller instructed the pilot to turn 10 degrees left from his current heading of 030 degrees. The pilot responded that he was turning to heading 040 degrees, but the controller did not correct the incorrect readback. The flight was given several more heading changes, and the controller informed the pilot to expect to be turned onto the downwind leg. At 1008:08, the controller vectored the pilot to fly heading 220 degrees, which the pilot acknowledged.

Air traffic control communications were transferred to the Boston Consolidated TRACON Final 1 position and on the pilot’s first contact at 1008:30, he informed the controller he was turning to heading 210 degrees, though approximately 22 seconds earlier he was previously instructed to fly heading 220 degrees. Approximately 1 minute 49 seconds later the controller instructed the pilot to turn 10 degrees to the right and descend and maintain 4,000 feet. The pilot acknowledged the heading and altitude clearances and radar data depicted the airplane flying the heading and descending with a ground speed of approximately 170 knots. At 1013:01, when the flight was approximately 21 nautical miles southwest of BOS flying at 4,200 feet msl, the controller instructed the pilot to turn to heading 170 degrees. The pilot acknowledged and radar data depicted the heading change.

The transcription of communications further indicates that at 1013:42, the controller instructed the pilot to turn to heading 130 degrees and then instructed him to descend and maintain 3,000 feet msl; the pilot acknowledged both clearances and recorded radar data reflects the airplane flying in a southeasterly heading. At 1014:44, when the flight was about 3,500 feet msl and approximately 1.5 statute miles west of the ILS 4R localizer at 170 knots ground speed, the controller instructed the pilot to turn left heading 060 degrees and to intercept the localizer. The pilot acknowledged the instructions and recorded radar data indicates a large radius turn to the left but the pilot flew thru the localizer flying in a east-northeasterly direction. At 1015:46, while flying at 3,300 feet msl on a heading of approximately 072 degrees approximately 1.6 statute miles east of the ILS 4R localizer, the controller instructed the pilot, “…you’re passing through the localizer turn left heading zero one zero intercept localizer.” The pilot responded with "one five Delta." After the pilot's truncated transmission, radar data indicated the flight turned left to a heading of approximately 320 degrees and descended to 2,800 feet msl then climbed to 2,900 feet msl. The recorded radar data depicted that between 1016:27, and 1016:39, the flight descended to 2,800 feet msl and turned to the right flying in a northerly direction. At 1016:39, the controller asked the pilot if he was joining the localizer; there was no response. At 1016:46, the controller instructed the pilot that the altitude indicated 2,300 feet, then immediately informed him radar contact was lost and ended the transmission with the partial call sign of the flight. The pilot responded with, "one five Delta with you."

The recorded radar data indicates that between 1016:39, and 1016:54, the airplane turned to the right heading in a southeasterly direction and descended to 1,900 feet msl. At 1016:57, the controller instructed the pilot that the flight was at 1,200 feet, issued a low altitude alert, and instructed the pilot twice to "climb immediately." The pilot responded "one five Delta’s climbing", and the recorded radar data indicates the flight turned to the left flying in a northwesterly direction and climbed to 2,200 feet msl. The controller advised the pilot to maintain 3,000 feet; and to state his heading. While the transcription of communications indicates the transmission from the pilot was unintelligible, review of the voice communication tape by the National Transportation Safety Board (NTSB) revealed the pilot responded, “15 Delta’s ah 020.”

The recorded radar data indicated that at 1017:27, the flight was at 2,700 feet msl and approximately 4 seconds later at 1017:31, the airplane was at 2,100 feet msl. The controller again instructed the pilot to maintain 3,000 feet and to state heading; there was no response. The controller then instructed the pilot that, "… your altitude is going up and down are you all right sir." There was no response. The controller then issued a low altitude alert again, instructed the pilot to climb immediately, and that radar contact was lost. The controller asked nearby aircraft if they heard a signal from an emergency locator transmitter; no signal was heard.

A witness who was on Route 106 reported to law enforcement seeing the airplane climbing and descending from the clouds.

A pilot-rated witness who was located .41 nautical mile north-northwest from the crash site reported hearing a low flying airplane, heading west to east, with high rpm. He ran outside but did not see the airplane. The noise faded then returned, and he observed the airplane descending vertically out of the clouds. The witness's first view of the airplane was the top of the airplane. He did not recall if he saw the landing gear, but did not see any smoke trailing the airplane during the time he saw it. The airplane spun 2.5 times to the left, went out of sight behind trees, and he heard an impact and saw smoke. He drove to the scene arriving there about 1 minute later. The witness added that the engine was running with a constant sound prior to impact, and that there was no missing or sputtering. The weather conditions at the time consisted of a ceiling at 800 feet, with no rain or wind.

One witness located adjacent to the accident site reported hearing the airplane then seeing it flying in a southerly direction at the base of or partially in the clouds. The airplane climbed into the clouds then heard it flying eastbound. About 15 to 20 seconds later he heard “coughing sounds,” followed by seeing the airplane emerge from the base of the clouds with no engine no operating. He then reported the engine began operating but was sputtering while the airplane was descending with the nose at an angle of about 45 degrees. He noted the airplane was turning clockwise at full throttle then abruptly began rotating counterclockwise. The witness did not see the impact but reported the engine was operating at high rpm before hearing the impact. He called 911 to report the accident, and reported that the cloud bases were low like hanging fog but there was no rain.

A witness who was located on the roof of a nearby store reported to law enforcement seeing the airplane flying low between their location and another nearby building. Numerous witnesses also reported to local law enforcement seeing the airplane circling and then nose diving into the parking lot.

The airplane impacted onto a parking lot during daylight conditions; the crash site was located approximately 212 degrees and 21 nautical miles from the center of BOS. There were no ground injuries or damage to any vehicles or equipment located in the parking lot.


The pilot, age 65, held a commercial pilot certificate with ratings airplane single engine land, airplane multi-engine land, and instrument airplane, issued September 4, 2003, and a second class medical certificate issued October 30, 2007, with a restriction to wear corrective lenses for near and distant vision.

Review of the pilot’s FAA certified medical file and also personal medical records obtained with consent from the pilot’s family was performed by the NTSB's Medical Officer. The review of the medical file and personal medical records revealed an FAA form 8500-14 (Ophthalmological Evaluation for Glaucoma) dated August 7, 2007, noted that the pilot had glaucoma treated with laser surgery and 3 different eyedrops, “trace” cataracts, and visual field deficits. Visual acuity on that date was noted (with correction) as 20/20 in the right eye and 20/20+2 in the left eye. The pilot’s Application for (2nd class) Airman Medical Certificate dated October 30, 2007, noted the glaucoma and treatment, and the issuance of a 2nd class medical certificate, without any further indication of additional evaluation. Ophthalmologist records were reviewed; they noted on February 15, 2008, that the pilot indicated his distant visual acuity was “not as clear as it used to be” and on May 13, 2008, the pilot’s visual acuity was noted to be “stable both eyes,” measured (with correction) at 20/20-2 and 20/25-2. Visual field examinations dated through February 15, 2008, documented persistent visual field deficits primarily in the upper half of the pilot’s visual fields in both eyes. The pilot’s personal medical records also documented an incident of several weeks of left sided facial numbness through December 18, 2008, though the pilot’s family reported no recollection of him complaining of any facial numbness.

NTSB review of the pilot’s 5th pilot logbook that begins with a “From” date of “1/1/2004” to the last entry dated July 28, 2008 (15 days before the accident), revealed a carry forward time of 1,408 hours. His family reported that he did not fly between the last logged flight and the accident date. Between the “From” date and the last entry in the 5th logbook, he logged approximately 113 hours total time, 12.8 hours as flight in actual instrument conditions, and executed 16 instrument approach procedures as pilot-in-command (PIC), all of which were in the accident airplane. His last flight review in accordance with 14 CFR Part 61.56 occurred on August 21, 2006. There was only 1 logged flight in which the pilot obtained simulated instrument instruction. The instruction occurred during the pilot’s flight review on August 7, 2004, which included three instrument approaches and .8 hour simulated instrument flight.

Further review of his No. 5 logbook for the previous 24 calendar months (August 1, 2006 to the accident date) revealed he logged 5.3 hours actual instrument flight time and seven instrument approaches as PIC. The dates of the instrument flights in the previous 24 calendar months that concluded with instrument approach procedures performed as PIC were August 4, 2006, February 3, 2008, and June 23, 2008. No logged actual or simulated instrument flight occurred between August 4, 2006, and February 3, 2008. Five instrument landing system (ILS) approaches were performed as PIC on June 23, 2008. A review of logbook entries in 12 calendar month intervals beginning with May 2004 (May 1, 2004 thru May 31, 2005), June 2005 (June 1, 2005 thru June 30, 2006), July 2006 (July 1, 2006 thru July 31, 2007), and August 2007 (August 1, 2007 thru the accident date), revealed he logged executing as PIC 4, 5, 1, and 6 instrument approach procedures respectively. There was no record that an instrument proficiency check was performed in the No. 5 logbook.


The airplane was manufactured in September 1956, by Beech Aircraft Corporation, as model G35, and was designated serial number D4807. It was powered by a Teledyne Continental Motors E-225-8 engine, and equipped with a Beech 215-107 constant speed propeller. It was also equipped with a single axis (roll) autopilot control system, and yaw damper.

Review of the maintenance records revealed the airplane was last inspected in accordance with (IAW) an annual inspection on August 9, 2007; the airplane total time at the time of the inspection was 3,739.94 hours. Based on the pilot’s logbook, he had operated the airplane for 30.3 hours since the last annual inspection. The pitot static system, altimeter, pressure altitude reporting system and transponder checks were last performed IAW 14 CFR Part 91.411 and 14 CFR Part 91.413 on October 23, 2006.

While the pilot documented airplane discrepancies in the remarks section of his 5th pilot logbook, a review of it revealed no entry documenting a discrepancy or malfunction with the primary or secondary flight controls, autopilot, yaw damper, flight instruments, or vacuum system.


A surface weather observation taken at BOS at 1015, or approximately 2 minutes before the accident, indicated the wind was from 350 degrees at 8 knots with gusts to 14 knots, the visibility was 8 statute miles with light rain, broken clouds existed at 1,100 and 2,000 feet, overcast clouds existed at 5,500 feet, the temperature and dew point were 17 and 15 degrees Celsius respectively, and the altimeter setting was 29.70 inches of mercury.


According to FAA personnel, there were no abnormalities with either the 4R Localizer or Glide Slope facilities on the accident date. The full ILS was operational that day. The Localizer and Glide Slope had a Periodic with monitors flight inspection on March 9, 2008. There was also a special flight inspection of the Localizer on April 28, 2008, for the purpose of an Expanded Service Volume authorization. Additionally, there were no unscheduled outages or abnormalities for the Glide Slope in the prior 12 month period; only scheduled maintenance was performed.

The Localizer had two reported incidents, none of which was associated with equipment performance abnormality. The first incident was associated with a snow outage; which was resolved by snow removal from the Localizer antenna's on January 14, 2008. The second incident was associated with a telephone line outage for the interlock signal. This incident was resolved on October 12, 2007, by restoration of a T-1 line from the Boston Air Traffic Control Tower (Boston ATCT) to the Boston Localizer (in Winthrop, MA) restoring the interlock signal to the facility.


The pilot was in contact with Boston Consolidated TRACON at the time of the accident; there were no reported communication difficulties.


BOS is equipped in part with runway 4R, which has an ILS or localizer approach. The approach requires radar or distance measuring equipment (DME).


Examination of the accident site revealed the airplane came to rest upright on a magnetic heading of 254 degrees. A ground scar associated with the right wing was noted just forward of the resting point of the right wing, the ground scar was oriented on a magnetic heading of 206 degrees. An impact crater associated with the engine was noted in the area where the airplane came to rest, and a propeller blade counterweight was found inside the impact crater. Additionally, a ground scar measuring approximately 34 inches long and parallel to the right wing ground scar was noted in the area of the impact crater. A second ground scar measuring approximately 17 inches in length was noted across the first ground scar from the engine impact crater forming an “X.” One propeller blade was found in a wooded area forward of the resting point of the wreckage, while the second propeller blade was found in the main wreckage area.

Examination of the wreckage revealed the cockpit, cabin, and portions of both wings were consumed by a postcrash fire. All components necessary to sustain flight were attached or found in close proximity to the main wreckage. No evidence of preimpact failure or malfunction was noted of the primary flight controls for roll, pitch, or yaw. The elevator trim actuator was extended 1 inch which equates to 0 degree trailing edge tab deflection. Examination of the autopilot roll servo revealed both ends of the bridle cable remained connected to the aileron balance cable, and the bridle cable remained wrapped around the capstan, which was found to rotate freely. A component of the autopilot system called the Radio Coupler was found forward of the resting location of the airplane. The switch on the Radio Coupler was found positioned to the “HDG” or heading position. The autopilot roll servo and Radio Coupler were retained for further examination. The flaps and landing gear were in the retracted position.

Examination of the cockpit revealed extensive impact and heat damage. The directional gyro (DG), which is one of two flight instruments that provide sensor information to the autopilot system was separated from the instrument panel and exhibited extensive heat and impact damage. Examination of the rotor and rotor housing from the DG revealed scoring at the end of the rotor housing. The impact damaged No. 1 VOR/Localizer Converter and Glide Slope indicator, and the heat damaged No. 1 communication and navigation transceiver were retained for further examination. The pilot’s pilot logbook and numerous navigation charts were found in the cockpit.

Examination of the engine and engine accessories revealed extensive impact and fire damage which precluded operational testing. The forward portion of the crankcase including the No. 6 cylinder, the crankshaft forward of the No. 5 journal, and the propeller hub were separated and located approximately eight feet from the main wreckage. No evidence of preimpact failure or malfunction was noted to the fracture surface of the crankshaft. Both magnetos, engine-driven fuel pump, carburetor, oil pump, starter, generator, and vacuum pump were separated from the engine but found within the main wreckage area. Examination of the engine assembly revealed no evidence of preimpact failure or malfunction. Examination of the vacuum pump revealed the drive gear exhibited impact damage and could only be turned by hand through 270 degrees of rotation. Disassembly of the vacuum pump revealed the rotor and rotor vanes were not failed. Examination of the separated engine-driven fuel pump revealed the drive shaft was fractured near the fuel pump housing and was not recovered or located in the main wreckage. Disassembly of the pump revealed the rotor and rotor vanes did not reveal any preimpact failures.

Examination of the propeller hub which remained attached to the crankshaft flange revealed both propeller blades were separated. Examination of the propeller blade that was found in the main wreckage area revealed leading and trailing edge damage. Chordwise scratches, aft bending, and leading edge blade twisting were noted from mid span to the blade tip. Examination of the propeller blade that was found forward of the resting location of the main wreckage revealed chordwise and spanwise scratches on the cambered side of the blade, and it was bent aft slightly.


Postmortem examination of the pilot was performed by the Office of the Chief Medical Examiner. The cause of death was listed as “Blunt impact of head, torso, and extremities with extensive skeletal and visceral injuries.” Postmortem examinations of the passengers were not performed; only external examinations.

Forensic toxicology was performed on specimens of the pilot by the FAA Bioaeronautical Sciences Research Laboratory, Oklahoma City, Oklahoma. The toxicology report stated the results were negative for carbon monoxide and tested drugs. Testing for cyanide was not performed, and the results were positive for ethanol in the blood (28mg/dL), liver (29 mg/dL) and muscle (16 mg/dL). A data field of the FAA Toxicology report titled Putrefaction was marked with “Yes.” Testing of specimens of the pilot was also performed by the UMASS Memorial Medical Center. The result was positive in the submitted blood specimen for ethanol (0.02 gm%). The result was negative for drugs of abuse screen, and also the alkaline drug screen.


Review of 14 CFR Part 61.57 revealed that no person may act as PIC under IFR or in weather conditions less than the minimums prescribed for visual flight rules (VFR), unless within the preceding 6 calendar months, that person has performed and logged under actual or simulated instrument conditions, either in flight in the appropriate category of aircraft for the instrument privileges sought or in a flight simulator or flight training device that is representative of the aircraft category for the instrument privileges sought: (i) at least six instrument approaches; (ii) holding procedures; and (iii) intercepting and tracking courses through the use of navigation systems. The regulation further states that a pilot who does not meet the instrument experience requirements within the prescribed time, or within 6 calendar months after the prescribed time, may not serve as PIC under IFR or in weather conditions less than the minimums prescribed for VFR until that person passes an instrument proficiency check consisting of a representative number of tasks required by the instrument rating practical test.

Examination of the KX155 Navigation and Communication transceiver and KI-209 VOR/Localizer Converter and Glide Slope indicator were performed at the manufacturer’s facility with FAA oversight. The examination of the transceiver revealed extensive heat damage which precluded operational testing and also determination of communication and navigation active and standby frequencies. Impact damage and missing components of the indicator precluded operational testing.

The installed Century IIB autopilot system utilizes a lighted directional gyro with a heading bug and a lighted attitude gyro for its sensing elements. These instruments work with a solid-state computer, control console and roll servo to provide lateral stabilization, roll command, and heading select. Preflight action requires a pilot to confirm the ability to override the autopilot using the control wheel. In normal operation the maximum bank in heading or “HDG” mode is 20 degrees.

Examination of the Radio Coupler and autopilot roll servo was performed at the manufacturer’s facility with NTSB oversight. Visual inspection of the exterior of the Radio Coupler revealed slight impact damage, while internal inspection revealed the top circuit board was impact damaged resulting in rearward tilting of two capacitors on the top board, and a capacitor on the bottom circuit board was loose. New connectors were installed and the unit was operationally tested. The unit passed all tests with the exception of the Step “2” procedure which tests the lag time for the Radio Coupler output reading to decay 40 percent within 3 to 6 seconds after the radio signal is turned off. The Step “2” failure was attributed to the capacitor on the lower circuit board being loose and not connected in the circuit. The roll servo was not able to be tested due to the extensive fire damage.

Radar data from the Providence ASR-9 radar system was plotted by NTSB personnel. The radar plots revealed the accident airplane was the sixth in approach/landing sequence for runway 4R. Further review of the radar plots revealed about the time the accident airplane was on runway 4R final approach, an Embraer 145 was approximately 4 miles ahead and a Boeing 757 was approximately 9 miles ahead. Federal Aviation Administration Order 7110.65P, titled Air Traffic Control, requires 5 miles separation of a small aircraft behind a heavy aircraft (Boeing 757).

Air Charity Network (ACN) is comprised of seven independent member organizations, of which Angel Flight Northeast is one of the seven member organizations. According to their (ACN) web site, they provide access for people in need seeking free air transportation to specialized health care facilities or distant destinations due to family, community or national crisis. The criteria to help an individual(s) are that they must be medically stable, ambulatory, and will not require medical care while en route. Pilots and aircraft owners volunteer their time and aircraft including the cost for fuel while operating IAW 14 CFR Part 91. Air Charity Network holds yearly meetings with the member organizations and disseminates information but does not dictate policy or procedure.

Angel Flight Northeast became incorporated on March 20, 1996, and the first flight occurred on May 31, 1996. The president and founder of Angel Flight Northeast reported that the accident flight was the first accident since incorporating. He also stated that they have 900 active pilots, and they operate on a system of trust; Angel Flight Northeast personnel do not perform checkrides with volunteer pilots; however, every year they obtain copies of a pilot’s medical certificate, pilot logbook, and insurance certificate. They review the pilot’s pilot logbook to check for cross country flight time and a current flight review. Angel Flight Northeast considers the pilot to be the operator, and as such they do not address or get involved in maintenance issues. Additionally, they do not have the capability to keep track of a pilot’s instrument currency, nor do they have a pilot complete a form for every flight documenting currency, or instrument currency. Angel Flight Northeast’s Pilot Manual in effect at the time of the accident specifies that pilot applicants must hold a private pilot certificate with an instrument rating, have a minimum of 250 hours as PIC, 25 hours in category and class aircraft to be flown for the mission, have a current medical certificate, and be current for both VFR and IFR flights.

The accident pilot submitted an application to Angel Flight Northeast dated November 1, 2007, which was received 5 days later. He received an orientation on November 20, 2007, and he logged trial run flights on January 13, 2008, lasting a total of 1.7 hours. An individual from Angel Flight Northeast flew with the accident pilot on the first three flights for the purpose of making sure he was completing the paperwork correctly. The accident pilot signed up for 20 flights, but actually flew 11 flights excluding the accident flight. Since October 7, 2004, the passengers had scheduled 72 flights thru Angel Flight Northeast, but flew 38 flights excluding the accident flight.

Use your browsers 'back' function to return to synopsis
Return to Query Page