N412ESBell 412EP · 2006

On November 24, 2008, about 1045 central standard time, a Beech 95-B55, N412ES, registered to and operated by a private individual, experienced an in-flight loss of control and crashed into a wooded area behind a house in Whites Creek, Tennessee. 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 business flight from Memorial Field Airport (HOT), Hot Springs, Arkansas, to Nashville International Airport (BNA), Nashville, Tennessee. The airplane was destroyed by impact and a postcrash fire and the certificated commercial pilot and 2 passengers were killed. The flight originated about 0909, from HOT.

Before takeoff, the pilot contacted the Memphis Air Route Traffic Control Center and requested IFR clearance to the destination airport. The pilot was instructed that the flight was cleared as filed, to climb and maintain 5,000 feet, expect 7,000 feet ten minutes after departure, instructed as to the discrete transponder code (5524), and advised of the clearance void time. The pilot correctly read back the clearance, and at 0911, he established contact with the facility and advised the controller that the flight was climbing through 2,500 feet. The flight was radar identified and instructed to climb to 7,000 feet continuing towards the destination airport.

Air traffic control (ATC) communications were transferred to several ATC facilities and at approximately 1024, while in contact with the Memphis Air Route Traffic Control Center, the pilot was instructed to descend and maintain 4,000 feet. At approximately 1032, ATC communications were transferred to the Nashville Air Traffic Control Tower, and at 1033:12 while in contact with that facility, the pilot advised the controller that the flight was at 4,000 feet. Radar data indicated that the pilot maintained approximately 4,000 feet from 1032:53 until 1044:12. The controller acknowledged the pilot contact and advised him that the automated terminal information service (ATIS) sierra was current and to expect vectors for an ILS approach to runway 20R at the destination airport. The pilot correctly read back the ILS runway designator and approximately 3 minutes later, at 1036:24, the pilot was instructed to fly heading 055 degrees for sequencing, which he also correctly read back. The radar data indicated that from 1037:21, to 1042:21, the airplane proceeded on the approximate heading instructed by the controller, but beginning at 1042:30, heading deviation left of the assigned course was noted.

At 1042:58, the controller advised the pilot of the location of another airplane and the pilot responded at 1043:05, "yeah well I’m all screwed up up here so stick with me." The controller asked the pilot if he needed assistance and the pilot asked the controller how far he was from the localizer. The controller advised the pilot that the flight was 12 miles from the localizer and asked the pilot to make sure he wasn’t confusing the John Tune Airport localizer setting (110.3 MHz) with the BNA Airport runway 20R localizer setting of 111.3 MHz. While the transcription of communications indicates the pilot responded at 1043:43 with, "(unintelligible) again one eleven three", Safety Board review of the voice tape from the facility revealed the pilot replied with words to the effect that he had 111.3 selected.

The controller then instructed the pilot to fly his present heading and to expect a base and final turn to the localizer; the pilot did not respond. The controller again advised the pilot to expect a base turn in 5 miles and a turn south onto the localizer. The transcription of communications indicates the pilot replied "gotcha." At 1044:24, the controller repeated the partial call sign of the airplane, and the pilot responded, "two echo sierra I’ve got you but I'm having trouble hold." A pause was noted and then grunting sounds were recorded on the frequency. The controller asked the pilot if he needed help and he responded, "I got it into a spin and I can’t stop it." Heavy breathing/grunting sounds for several seconds were heard on the frequency. At 1044:54, the controller advised the pilot to climb immediately to 3,000 feet. The pilot responded with "got no climb." At 1045:09, the controller again advised the pilot to climb immediately; there was no further response from the pilot. The controller broadcast on the frequency advising the pilot that the flight was 6 miles north of John Tune Airport, and that rescue vehicles had been dispatched.

According to the controller who was communicating with the accident pilot, he noticed the pilot was not tracking his assigned heading, and then noticed a descent from the assigned altitude below the minimum vectoring altitude. He issued climb instructions and reported the accident airplane kept descending. He marked the aircraft's position and asked to be relieved from his position.

Radar data correlated with the transcriptions of communications revealed that at 1042:30 which was approximately 35 seconds before the pilot first advised the controller of a discrepancy, the airplane's heading changed to the left without instruction from ATC to a maximum northerly heading of 023 degrees which occurred at 1043:03. The airplane then turned to 030 degrees and remained on that heading for 3 radar returns between 1043:08 and 1043:17. Between 1043:21 and 1043:31, the airplane turned to the right flying the approximate heading instructed previously by the controller (055 degrees), and between 1043:31 and 1044:12, heading changes to the left and right were noted as well as a decrease in groundspeed to 155 knots. Between 1044:12, and 1044:27 which was the approximate time the pilot advised the controller that he was "…having trouble hold", the radar data indicates that the airplane's heading changed from 051 degrees to 344 degrees, the altitude decreased from 4,000 to 3,700, and the groundspeed decreased from 155 to 145 knots.

Further review of the radar data correlated with the transcription of communications revealed that between 1044:27, and 1044:40 which was the time the pilot advised the controller that the airplane was in a spin and couldn't stop it, the radar data reflected that the heading changed from 344 degrees to 035 degrees, altitude deviations between 3,700 and 3,500 were noted, and the groundspeed decreased from 139 to 107 knots. Between 1044:40 and 1045:03 (last correlated radar return), heading changes left and right were noted, the altitude decreased from 3,500 to 1,300 feet, and the groundspeed decreased from 107 to 51 knots. The last radar target was located at 036 degrees 17 minutes 22.01 seconds North latitude and 086 degrees 50 minutes 38.38 seconds West longitude, at an altitude of 1,300 feet. The accident site was located approximately 52 degrees and 0.04 nautical mile from the last radar return.

A witness located near the crash site reported he was inside his house and heard an airplane that sounded abnormal; the witness and his son reported hearing a whirring sound. The witness looked outside his window and saw the airplane flying in a southwesterly direction. The abnormal sound faded momentarily, and then returned. The airplane began spinning in a counterclockwise direction and was flat while spinning. He estimated the airplane spun 4 times before losing sight. Each 360 degree rotation took approximately 1 second. He called 911 and ran to the scene. When he arrived there he noticed a fire at the front of the airplane which spread out. He estimated the fire department arrived in 5 minutes. The witness further reported the weather condition at the time consisted of low clouds with light drizzle.

The pilot, age 67, held a commercial pilot certificate with airplane single engine land, airplane multi-engine land, and instrument airplane issued June 28, 2002. His most recent

special issuance third class medical certificate was issued November 21, 2008, with limitations to have available glasses for near vision.

The pilot reportedly kept his pilot logbook inside the cockpit; the logbook was not located during the inspection of the wreckage. He listed 1,990 hours of total flight time on the application for his latest medical certificate. While no pilot logbook was located, correlation of the pilot's flight time based on maintenance record entries revealed the airplane had been operated approximately 20 hours between November 1, 2007, and May 26, 2008 which was the last entry in the engine logbooks.

The pilot was paid on a monthly basis to fly the passengers on business trips when needed. During those trips, the pilot's fuel costs, lodging and meals were also paid for by the passenger's company.

The airplane was manufactured in 1978, by Beech Aircraft Corporation, as model 95-B55, and was designated serial number TC-2198. It was powered by two Teledyne Continental Motors IO-520-E engines rated at 260 horsepower and 2,700 rpm for maximum continuous operation that were installed in accordance with (IAW) Supplemental Type Certificate (STC) SA432SO. The flight manual supplement (FMS) associated with the STC indicated the normal operating (green arc) range was 2,000 to 2,700 rpm. It was also equipped with two Hartzell HC-EHC-G3YF-2UF propellers, which are single-acting, hydraulically operated, constant speed with manual feathering capability.

Review of the airframe maintenance records revealed the last entry pertaining to the transponder, altimeter, and pitot static system tests in accordance with 14 CFR Part 91.411, and 91.413 was logged as occurring on June 24, 2003. The work was performed by a Federal Aviation Administration (FAA) approved certified repair station. While there was no further entry in the airframe logbook indicating the transponder, altimeter, and pitot static systems were checked after that date, an individual who knew the pilot and the pilot's wife reported that the checks were performed on November 22, 2006. The individual who knew the pilot also reported that the facility that performed the work is no longer in business. The airplane was last inspected in accordance with an annual inspection on November 1, 2007; the airplane total time at that time was recorded to be 2,802.7 hours. The inspection entry indicates that IFR certification was overdue.

The left and right engines were remanufactured at the Teledyne Continental Motors factory on February 3, 2008, and February 2, 2008, respectively. Both were shipped to a distributor (Aviall) on February 3, 2008.

Further review of the maintenance records revealed that on February 29, 2008, both propellers and propeller governors were removed for overhaul, and both engines were removed and replaced with remanufactured engines. The airplane total time at that time was recorded to be 2,807.7 hours. The entry does not indicate that new flexible fuel hoses were installed. The mechanic who performed the installation of the engines stated that he pressure tested the fuel hoses and if they did not leak he re-used them. Entries in the left and right engine logbooks revealed that on May 26, 2008, at airplane total time 2,832.9 hours, the same mechanic who installed the engines changed the engine oil and cleaned the oil screens of both engines. The entry for the left engine indicates, "…Ground Run. No Leaks Noted At This Time." There was no further entry in the left engine logbook following the oil change indicating any further maintenance was performed.

The airplane was modified by a FAA certified repair station on November 13, 2008, IAW STC SA09411AC-D. The STC removed the previously installed autopilot components and installed an S-Tec System 30 autopilot. Review of the FAA Approved Pilot's Operating Handbook and/or Airplane Flight Manual Supplement revealed that in the event of single engine operations during cruise flight, the procedures specify to retrim the airplane, and perform normal engine out procedures. The removal of the previous autopilot components and installation of the S-Tec autopilot system was not recorded in the permanent maintenance records; however, a logbook entry sticker documenting the work was later provided to Safety Board personnel. Review of the work order and logbook entry revealed no documentation of the airplane total time.

The day before the accident, about 1917, the pilot called the Lockheed Martin Automated Flight Service Station and filed 2 IFR flight plans and was given standard preflight weather briefings. The first briefing and flight plan was for the intended flight from HOT to BNA, and the second briefing and flight plan was for another flight that same day from BNA to Asheville, North Carolina.

According to the transcription of the weather briefing for the flight from HOT to BNA, the pilot was advised that a cold front was expected to move into the area of HOT about mid-morning, with the frontal boundary located just west of Nashville with rain showers or rain showers type activity ahead of the front. The terminal area forecast for the destination airport up until noon was for the visibility to be occasionally 2 miles with rain showers, ceiling 700 feet overcast clouds. The briefing specialist advised the pilot that the freezing level for the destination airport area was forecast to be at 9,000 feet, and provided him with the winds aloft.

A special surface weather observation taken about 1036 at BNA on the day of the accident, or approximately 9 minutes before the accident, indicated the wind was from 210 degrees at 10 knots with gusts to 16 knots; the visibility was 3 miles with moderate rain and mist; scattered clouds existed at 1,600 feet, broken clouds (ceiling) existed at 2,200 feet, and overcast clouds existed at 4,500 feet. The temperature and dew point were 52 degrees and 48 degrees F respectively, and the altimeter setting was 29.98 inches of mercury (inHg).

According to the transcription of communications with the Nashville Airport Air Traffic Control Tower, at 1038:39, during an air traffic control turnover briefing, local weather was briefed to the new controller who was informed there were no icing reports. The report of cloud bases at 1,200 feet exactly 1 hour earlier by an airplane at a nearby airport was also briefed.

A surface weather observation taken about 1053 at BNA on the day of the accident, or approximately 8 minutes after the accident, indicated the wind was from 200 degrees at 9 knots with gusts to 16 knots; the visibility was 3 miles with light rain, and mist; broken clouds (ceiling) existed at 1,800 feet, and overcast clouds existed at 2,600 feet. The temperature and dew point were 52 and 48 degrees F; and the altimeter setting was 29.97 inHg.

Safety Board review of the BNA Daily Record of Facility Operation for the day of the accident revealed no navigational aid outages related to the ILS for runway 20R.

No communication problems were reported.

Examination of the accident site revealed the airplane came to rest upright on a magnetic heading of 150 degrees in a wooded area behind a residential area. The accident site was located at 36 degrees 17.393 minutes North latitude and 086 degrees 50.597 minutes West longitude, at an elevation of 646 feet, and was also located approximately 13 nautical miles and 321 degrees from the center of BNA.

Further examination of the accident site revealed the airplane impacted on a 10 to 15 degree downsloping terrain of a wooded area. Fire damage to a small area surrounding the wreckage was noted. The surrounding tree heights were approximately 60 feet and there was no evidence of a swath through the trees. Numerous trees surrounded the wreckage were noted. One approximately 60 foot tall tree was located immediately forward of the left wing outboard of the engine nacelle, and one 60 foot tall tree was located about midspan of the left flap. One approximately 45 foot tall tree was located at the right leading edge outboard of the engine nacelle, and one approximately 57 foot tall tree was located at the aft edge of the right flap near the outboard portion of the flap. The tree located forward of the left wing exhibited a curved scrape mark approximately 13 feet above ground level, and the tree located by the rear of the left wing exhibited scraping of the east side of the trunk. The tree located at the trailing edge of the right wing was broken approximately 17 feet above ground level and all limbs were broken from the ground up to approximately 10 feet above ground level.

Examination of the airplane revealed a postcrash fire consumed the cockpit, cabin, and sections of both wings. Fire damage also consumed the empennage from the horizontal stabilizer forward to cockpit. Both wingtips, the ends of the horizontal stabilizers and also the upper portion of the vertical stabilizer remained attached. All primary flight controls remained attached with the exception of the left aileron, which was found approximately 15 feet forward of the main wreckage. Both engines remained attached to the airframe and both 3-bladed propellers remained attached to each engine.

Examination of the left wing revealed the fuel tank was destroyed by fire. The left wing which was comprised of 2 pieces was fractured at wing station (WS) 165, and the main spar was fractured at WS 140. The flap was attached at the inboard flap track and the outboard section of the flap was consumed by the postcrash fire. The aileron bellcrank was separated from the structure but the terminating ends remained attached to both aileron control cables. The aileron pushrod was bent but connected at the bellcrank, but fractured 9 inches outboard of the bellcrank attach point. The aileron trim tab was separated from the aileron; the trim tab was found approximately 15 feet away from the resting location of the left wing. The aileron trim tab actuator measured 1.10 inches extended, which equates to 8 degrees tab trailing edge up (wing up). The leading edge of the trim tab was crushed down. The aileron was separated and fractured into 2 pieces. The fracture point was 24 inches from the outboard end. The rod end was connected to the aileron and the pushrod was fractured at the end of the threaded rod end. The aileron was separated at the hinges; the counterweights were attached. The trailing edge was crushed forward and down which was consistent with tree contact. Pieces of tree were embedded in skin splice. The de-ice boot remained attached to the separated wing section; dirt was noted to be adhering to it. Approximately 31 inches of de-ice boot inboard of the landing light exhibited no fire damage. Tree contact (semi-circular indentation) was noted 21 inches inboard of the landing light. The boost pump was destroyed by fire, and the fuel strainer was heat damaged. The fuel screen was clean but the solder was melted and resolidified. The fuel selector remained attached by the mounting bracket and found positioned to the main tank, and the fuel selector supply lines were consumed by fire. The flap actuator was extended 1.75 inches, which equates to flap retracted, and the landing gear was retracted.

Examination of the right wing revealed the fuel tank was destroyed by fire; the fuel cap in place. Tree contact was noted at the leading edge at WS 119. The flap remained attached at the inboard flap track, and the outboard flap track was separated from the aft spar but remained attached to the flap. The full span of the aileron remained attached. The main spar was heat fractured at WS 114, and the aft spar was heat fractured at WS 86. The aileron rod end remained attached to the aileron, sections of aileron bellcrank remained attached to the aileron cables. Approximately 65 inches of de-ice boot remained attached to the outer portion of the wing; dirt and debris on the leading edge and upper surface of the de-ice boot was noted. The fuel strainer was heat damaged and the screen was clean. The fuel selector exhibited thermal and impact damage. The flap actuator was separated from aft spar and flap and found extended 1.626 inches which equates to flap retracted, and the landing gear was retracted.

Examination of the empennage revealed the rudder remained attached, and the rudder trim tab remained attached to the rudder. Both horizontal stabilizers, elevators, and elevator trim tabs remained attached. The left horizontal stabilizer exhibited impact damage to the lower skin panel mid-span, and the right elevator exhibited impact damage on the lower skin surface. The left and right elevator trim tab actuators were found extended 1.20 and 1.18 inches, which equates to 5 degrees tab trailing edge down (tail down). The rudder trim tab actuator was extended 3.74 inches, which equates to 5 degrees tab trailing edge right (tail right). No evidence of in-flight fire was noted on any structure, primary flight control, or secondary flight control surface.

Examination of the heat damaged cockpit revealed the throw-over control yoke was found lying in the cockpit with all flight control cables separated. A digital camera with extensive heat damage was noted; the camera was retained for further examination. Burned aviation terminal approach procedures documents were found on the left side of the aft baggage compartment. Remnants of baggage were found in nose baggage compartment; an estimated 30 pounds of baggage total was found in wreckage. The left and right fuel selector handle were each positioned to the main tank positions. Inspection of the landing gear actuator rod confirmed the landing gear to be retracted. The attitude indicator which was heat damaged was disassembled revealing slight rotational scoring to the rotor but no rotational scoring was noted inside the rotor housing. Flight and engine instruments were partially consumed and illegible. An engine Operator’s Manual and Pilot's Operating Handbook were found in the cockpit and cabin area. The positions of the cowl flap selectors and controls of the heat damaged throttle quadrant could not be determined. The pilot's lower escutcheon panel was separated and switch positions could not be determined.

Examination of the aft utility door revealed it separated from fuselage at the door hinge, and was partially consumed by the postcrash fire. The handle was found in the closed and latched position, and the pins were found extended. Examination of the cabin door revealed it was separated from the fuselage at door hinges by the postcrash fire, and was partially consumed by the postcrash fire. The main latch bolt was found extended and bent downwards. The outside door handle locking mechanism was missing.

Aileron flight control cable continuity was confirmed from the remains of the bellcrank near each control surface to the cockpit where the cables were found separated from the control yoke. Rudder cable continuity was confirmed from the control surface to the rudder control bar where the rudder pedals were consumed by fire, and elevator cable continuity was confirmed from the control surface to the cockpit.

Examination of the left engine revealed it was upright and remained partially attached to the airframe via various cables and lines. Two of the four engine mount legs appeared to be fractured. The engine and the engine accessories exhibited thermal damage, discoloration and debris. Various lines and hoses, the starter and starter adaptor, vacuum pump, fuel pump, fuel manifold valve and lines, the magnetos and ignition harness, as well as the crankcase and cylinders, except cylinder 5, exhibited thermal damage and / or thermal discoloration. The throttle was in the full open position, and the propeller governor control arm and mixture control arms were both found in the ¾ position of travel from their respective full (low pitch stop and full rich) positions. The oil cap was found securely fastened to the oil filler neck. The engine driven fuel pump exhibited thermal discoloration. The fuel supply line from the firewall to the engine-driven fuel pump was found disconnected from the fuel pump inlet fitting; the hose was not fractured. All of the inlet fitting threads exhibited thermal debris, and soot was noted on the interior of the inlet fitting. The fuel supply line “B nut” exhibited a white colored substance on the nut flats and sooting on the female threads of the nut. The engine driven fuel pump drive coupling was intact and undamaged; the fuel pump did not turn freely by hand. The fuel pump and fuel supply line were retained for further examination. The propeller blades were not in the feathered position; two of the three propeller blades did not appear to be bent, while the third blade was bent aft and was located under the engine. The cowl flap was observed to be in the open position.

Further examination of the left engine revealed the top spark plugs electrodes exhibited normal wear signatures. Inspection of the cylinders using a lighted borescope revealed light colored combustion deposits. Crosshatch hone marks were observed on all six cylinder walls. No anomalies were noted in the cylinder combustion chambers or valves. The magnetos and ignition harness exhibited thermal discoloration. The timing plug was removed from both magnetos and the internal components appeared intact. The left engine firewall exhibited severe thermal damage and sooting that was white in color. The outside of the left engine upper cowling exhibited thermal damage, paint bubbles, and buckling of the aft, right corner. The paint on the aft portion of the cowling appeared to be burned away. Soot was observed on the top, front “nose bowl” portion of the cowling, as well as the oil fill door area. The left engine and propeller were removed for further examination.

Examination of the right engine revealed it was upright and remained partially attached to the airframe via various cables and lines. Three of the four engine mount legs appeared to be fractured. The engine and the engine accessories exhibited thermal damage, discoloration and debris. Various lines and hoses, the starter and starter adaptor, vacuum pump, fuel pump, fuel manifold valve and lines, the magnetos and ignition harness, as well as the crankcase and all six cylinders exhibited thermal damage and / or thermal discoloration. The throttle was found in the full open position, and the propeller governor control arm and the mixture controls were each found at the ¾ forward range of travel. The oil cap was found securely fastened to the oil filler neck. The engine driven fuel pump which exhibited thermal discoloration was removed and the drive coupling was intact. The engine driven fuel pump rotated freely by hand.

Further examination of the right engine revealed the top spark plugs electrodes exhibited normal wear signatures. Inspection of the cylinders using a lighted borescope revealed light colored combustion deposits. Crosshatch hone marks were observed on all six cylinder walls. No anomalies were noted within the cylinder combustion chambers or valves. The magnetos exhibited thermal discoloration and the ignition harness was thermally damaged. The timing plug was removed from both magnetos and the internal components appeared intact. The vacuum pump exhibited thermal discoloration and the drive coupling exhibited thermal damage and melting. The vacuum pump was removed and would not rotate by hand. The vacuum pump aft cover was removed and the vanes and rotor were intact. Two of the six vanes could not be easily removed from the rotor. The vacuum pump cavity exhibited light scratches on the walls, consistent with normal operation. The retention plate between the vacuum pump cavity and the aft cover exhibited wear that was not uniform. The propeller blades were not in the feathered position. Two of the three blades did not appear to be bent, while the third blade was bent aft and was located under the engine. The cowl flap was observed to be in the closed position. The right engine firewall was sooted and exhibited thermal damage. The right engine upper cowling exhibited thermal damage and paint bubbling on the aft half of the cowling. The inside of the cowling exhibited soot and thermal damage throughout. The right engine and propeller were removed for further examination.

Examination of the left propeller revealed all three blades were at a low blade angle. Impressions in the spinner dome by the counterweights of the Nos. 2 and 3 propeller blades indicated they were at a low pitch position when the impact marks were made. The air valve was intact and still retained its charge. The low pitch stop had a very slight impression mark and the feather stop was intact and unremarkable. The start locks were intact and operable. A slight mark on the start lock weights was noted. No impact marks were noted on the preload plates for the Nos. 1 and 2 propeller blades, but the preload plate for the No. 3 propeller blade exhibited a contact mark made by the blade pitch change knob. The No. 1 propeller blade did not exhibit any bending or twisting, and there was no rotational scoring. The No. 2 propeller blade had a very slight abrasion on the trailing edge near the blade tip, but did not exhibit any bending or twisting. The No. 3 propeller blade was bent aft approximately 60 degrees at ¼ radius, and was also bent forward slightly at 2/3 radius. Mild spanwise scoring on cambered side of the blade near the leading edge and also spanwise scratches on the cambered side of the blade were noted. The blade back exhibited spanwise scoring in the blade tip area.

Examination of the right propeller revealed all three blades were at a high blade angle but not completely in the feathered position. An impression in the spinner dome from the No. 3 propeller blade counterweight indicated the blade was at a low pitch position when the impact mark was made. The air valve was intact and still retained its charge. The low pitch stop had a very slight impression mark and the feather stop was intact and unremarkable. The start lock was intact but all four retaining screws were stripped. An impact mark on both weights from the start lock sleeve was noted. Inspection of the preload plates for all three propeller blades revealed none exhibited evidence of impact marks that could be used to calculate propeller blade angle. The No. 1 propeller blade was not bent or twisted and did not exhibit any rotational scoring, but did exhibit localized areas of paint abrasion. The No. 2 propeller blade had a very slight forward bend and minor paint abrasion, but there was no evidence of rotational scoring. The No. 3 propeller blade was bent aft approximately 70 degrees at 8 inches from the blade butt. The outer 1/3 of the blade was mildly bent forward and was not twisted. Spanwise scoring was noted on the cambered side of the blade with some abrasion on the trailing edge noted.

Postmortem examinations of the pilot and passengers were performed by the Office of the State Medical Examiner, Nashville, Tennessee. The cause of death for all three was listed as, "Multiple Blunt Force Injuries."

Forensic toxicology was performed on specimens of the pilot by the FAA Bioaeronautical Sciences Research Laboratory, Oklahoma City, Oklahoma. Testing for carbon monoxide and cyanide was not performed due to an insufficient quantity of the submitted blood specimen. Unquantified amounts of irbesartan and metoprolol were detected in the submitted urine and liver specimens.

The medications found during toxicology testing had been reported on the pilot’s last application for medical certification and approved by the FAA for his use.

Forensic toxicology was performed on specimens of the passengers by Aegis Sciences Corporation (Aegis), Nashville, Tennessee. The result of analysis of specimens of the center row left seat occupant was positive for carbon monoxide (3 percent saturation) and negative for volatiles and tested drugs. The result of analysis of specimens of the center row right seat occupant was positive for carbon monoxide (10 percent saturation) and negative for volatiles and tested drugs.

With respect to the propeller, during normal operation the propeller governor increases oil pressure to the propeller cylinder which counteracts the forces of the spring, counterweight on each propeller blade, and an air charge attempting to rotate each propeller towards high pitch, or feather position. The oil pressure from the propeller governor is supplied to the propeller cylinder which in turn pushes against a piston rotating the blades from high pitch towards a low pitch setting. Absent oil pressure, the spring, counterweight on each propeller blade, and an air charge in the cylinder rotate each propeller blade to the high pitch or feathered position, though start locks prevent this from happening at the low rpm range such as during normal engine shutdown. Feathering of the propeller blades to reduce drag following an engine failure occurs by moving the cockpit propeller control to the feather position. The cockpit control is mechanically linked to the propeller governor which causes the propeller governor to dump oil pressure to the propeller cylinder, allowing the spring, counterweight on each propeller blade, and air charge in the cylinder to move the blades to the feather position.

The Pilot's Operating Handbook and FAA Approved Airplane Flight Manual (POH/AFM) indicates the following information and steps to be performed in the event of an engine failure after lift-off and in-flight:

The most important aspect of engine failure is the necessity to maintain lateral and directional control. If airspeed is below 78 knots, reduce power on the operative engine as required to maintain control.

An immediate landing is advisable regardless of take-off weight. Continued flight cannot be assured if take-off weight exceeds the weight determined from the TAKE-OFF

WEIGHT graph. Higher take-off weights will result in a loss of altitude while retracting the landing gear and feathering the propeller. Continued flight requires immediate pilot response to the following procedures.

1. Landing Gear and Flaps - UP

2. Throttle (inoperative engine) - CLOSED

3. Propeller (inoperative engine) - FEATHER

4. Power (operative engine) - AS REQUIRED

5. Airspeed - MAINTAIN SPEED AT ENGINE FAILURE (99 KTS MAX.) UNTIL OBSTACLES ARE CLEARED

After positive control of the airplane is established:

6. Secure inoperative engine:

a. Mixture Control – Idle CUT-OFF

b. Fuel Selector - OFF

c. Auxiliary Fuel Pump - OFF

d. Magneto/Start Switch - OFF

e. Generator/Alternator Switch - OFF

f. Cowl Flap - CLOSED

7. Electrical Load - MONITOR (Maximum load of 1.0 on remaining engine)

As previously reported in the Wreckage and Impact Information section of this report, the left propeller blades were not feathered and the left fuel selector valve was found positioned to the main tank position, both of which were contrary to the engine failure procedures specified in the POH/AFM.

Operational testing of both engines was performed at the manufacturer's facility. Heat and impact damaged components from both engines were removed and replaced with exemplar components, and a test club propeller appropriate for the engine make and model was installed during operational testing of both engines. During operational testing of the left engine, it was started and found to operate to 2,868 rpm. Subsequent operational testing of the left engine was performed with the B-nut at the engine-driven fuel pump supply inlet fitting at a value (finger tight), which is less than specified by the airframe manufacturer. Approximately 2 minutes into the engine run while operating at 2,300 rpm, the fuel hose separated from the inlet fitting. During the loosening process, fuel leakage was noted but the engine continued to operate at the set rpm setting until the hose completely separated. The engine was then secured for safety reasons. Additional engine runs were performed with the B-nut finger tight but the B-nut did not separate from the inlet fitting. During the testing, the B-nut was noted to stay in the same position throughout one operational test, while during another test the B-nut was noted to have loosened varying amounts at the completion of the engine runs. During the last operational test which lasted approximately 5 minutes with the engine operated at varying rpm settings, the B-nut was noted to loosen and tighten with changes in engine rpm. At the completion of the last operational test, the B-nut was noted to be 3 ½ flats loose from the finger tight position. Operational testing of the right engine revealed that it operated to maximum rpm of 2,932 though an oil leak at the No. 2 cylinder rocker boss precluded further operational testing of the engine.

A rebuilt engine-driven fuel pump with new fittings was installed onto the left engine at the time it was rebuilt. The flexible fuel supply line from the left firewall fitting to the engine-driven fuel pump inlet fitting is considered an airframe item, and was not included as a component of the rebuilt engine.

Review of the airframe maintenance records that contain an entry when the standard airworthiness certificate was issued (December 12, 1978) to the last entry dated February 29, 2008, revealed no record indicating the left flexible fuel supply hose part number MS28741-8-0184 was replaced. Further review of the maintenance records revealed that since the airplane was manufactured, the left engine was recorded as being removed on 3 separate occasions. The first entry dated May 5, 1990, at airplane total time 1,415.5 hours, indicates the airplane was modified by installation of 300 horsepower engines IAW STC SA432SO. The next entry dated August 26, 1991, at airplane total time 1,555.7 hours, indicates both engines were removed due to collapse of the nose landing gear. The final entry dated February 29, 2008, was for installation of the rebuilt engines which were installed at the time of the accident.

Hawker Beechcraft recommends that fuel hoses be replaced when conditions warrant, at engine overhaul, or 5 years from date of delivery, whichever occurs first.

Examination of the heat damaged engine-driven fuel pump and fuel supply line performed at the Safety Boards' Materials Laboratory revealed the fuel supply inlet fitting had a severely oxidized appearance and contained debris in the groves of the threads. The beveled sealing surface of the fitting had a circular witness mark indicating that at one time in its service history it had been fully seated against the sealing surface of a mating nut. Additionally, there was a large clump of soot adhering to the inner diameter of the fitting near the face, along with debris further down in the bore. The nut, which would have been secured to the fuel supply inlet fitting on the fuel pump, also exhibited severe oxidation and contained debris within the groves of the threads. An “out of round” appearance of the bore was attributed to soot accumulation and not due to a deformed component. After cleaning, the threads on the fitting of the pump and also on the fuel line were in good condition. The condition of the threads shows that the two mating components did not separate by force, which would have stripped or deformed the threads. There was no sign of cross threading, which could also have facilitated the separation of the two mating components. The two mating components were fitted together in the lab and assembled until the sealing surfaces were fully seated against each other. Assembling the two components together was smooth without any perceptible binding, and the sealing surfaces came into contact after 2 ¾ turns. Further examination of the fuel supply line opposite end which was found attached to its mating fitting on the firewall revealed it did not exhibit the same signs of severe oxidation and soot accumulation on the inner surfaces as does the one belonging to the fuel pump end, which was found disconnected. The damage sustained by the fuel supply inlet fitting on the fuel pump, and the corresponding nut on the fuel supply hose, is consistent with the inner and outer surfaces of these components having been exposed to the effects of a fire and therefore having been separate from each other during the fire.

Inspection of both propeller governors was performed at a FAA certified repair station with Safety Board oversight. The left propeller governor was placed on the test bench as received and operated to maximum travel of the control shaft. The maximum rpm noted was 2,890 (specification is 2,850 rpm). The unit was noted to pump 7 quarts per minute (specification is 6 quarts per minute minimum). The feather check was satisfactory and the governor dumped oil pressure at 2,100 rpm (specification is 2,100 plus or minus 25 rpm). The relief valve pressure was noted to be 410 pounds per-square-inch (specification is 320 psi). Disassembly of the unit following the bench testing revealed the relief valve spring measured 1.744 inches (specification is 1.760 inch plus or minus 0.020 inch. Inspection of the right propeller governor revealed the top cover and control shaft were bent which precluded operational testing. Additionally, slight heat damage was noted. Disassembly inspection of the governor revealed the flyweights and metering valve were satisfactory. Other than the impact and slight heat damage, there was no evidence of preimpact failure or malfunction.

Examination of the digital camera found in the wreckage by the Safety Board's Vehicle Recorders Division revealed all internal electronic components and circuit boards were massively heat damaged, many to the point of carbonization. None of the non-volatile memory (NVM) integrated circuits survived; therefore, no data could be recovered from the camera.

In September 2004, Raytheon Aircraft Company issued Safety Communique No. 249 to owners and operators of all Beech Baron airplanes. The communique reiterated that spin maneuvers are prohibited by the FAA for normal category twin-engine airplanes, and that a spin can occur whenever an airplane is stalled and is subject to yaw input from rudder, asymmetric power, aileron, p-factor, or any combination of these forces. The communique also reinforces that if a pilot recognizes the uncontrollable yaw, or experiences any symptom associated with a stall, the operating engine throttle should be sufficiently retarded to stop the yaw as the pitch attitude is decreased.

Acoustic analysis of sounds recorded during ATC communications from the pilot to the controller was performed by the Safety Board's Office of Research and Engineering. A total of nine communications from the pilot were selected for analysis. The first communication from the pilot to ATC advising the controller that the flight was at 4,000 feet indicates one engine rpm, which could be consistent with both engines operating in synchronized rpm. The rpm at the beginning of the communication was at 2,350 but had decreased at the end of the communication to 2,320. The second communication from the pilot to ATC indicating his acknowledgement to expect vectors for ILS approach to runway 20R indicates one engine rpm, which could be consistent with both engines operating in synchronized rpm. The rpm at the beginning of the communication was approximately 2,310, and was approximately 2,305 at the end of the communication. The third communication from the pilot to ATC indicating his read back of a vector indicates one engine rpm, which could be consistent with both engines operating in synchronized rpm. The rpm at the beginning and end points of the communication were both approximately 2,190. The fourth communication from the pilot to ATC indicating "yeah well I’m all screwed up up here so stick with me" indicates one engine rpm, which could be consistent with both engines operating in synchronized rpm. At the beginning of this communication the engine rpm was approximately 2,210 which increased to a maximum of 2,260, but at the end of the communication was at approximately 2,230. The fifth communication from the pilot to ATC indicating his question to the controller as the location from the localizer indicated one engine rpm, which could be consistent with both engines operating in synchronized rpm. The rpm at the beginning of the communication was approximately 2,230, increased to a maximum of approximately 2,265, then decreased at the end of the communication to approximately 2,140.

Further acoustic analysis of the recorded ATC communications from the pilot to ATC revealed that the sixth communication indicating his comment to the controller that he had the correct ILS frequency selected indicates one engine rpm signature, which could be consistent with both engines operating in synchronized rpm. The first signature was approximately 2,210 rpm, while the second signature began at 2,160 and decreased to 2,135 at the end of the communication. The seventh communication from the pilot to ATC indicating he was "…having trouble hold" indicates two engine rpm signatures. The first signature was approximately 2,130 at the beginning of the communication which decreased to approximately 2,070 at the end of the communication. The second signature was approximately 2,010 at the beginning of the communication and increased to a maximum of 2,040, but decreased to 2,020 at the end of the communication. The eighth communication from the pilot to ATC indicating he was in a spin indicates two engine rpm signatures. The first signature was approximately 2,150 at the beginning of the communication and decreased to 2,020 at the end of the communication. The second signature was 2,000 at the beginning of the communication and decreased to 1,780 rpm at the end of the communication. Additionally, the stall warning horn was recorded during the entire communication. The final communication analyzed (ninth) indicating his comment to the controller that he could not climb indicates two engine rpm signatures. The first signature was approximately 2,110 at the beginning of the communication and increased to 2,140 at the end of the communication. The second signature was approximately 1,800 at the beginning of the communication and decreased to and leveled off at 1,760 at the end of the communication. During the two transmissions when separate engine rpm signatures were noted, no determination could be made as to which engine was producing which trace.