Sunday, November 10, 2019

Abrupt Maneuver: Cirrus SR22T, N707DF; fatal accident occurred February 21, 2018 near Montgomery-Gibbs Executive Airport (KMYF), San Diego, California











Aviation Investigation Report: Loss of Engine Power due to Excessive Fuel Flow in Cirrus SR22T Aircraft 
The NTSB has released an Aviation Investigation Report (AIR-22-04) urging Cirrus Aircraft and the Federal Aviation Administration (FAA) to address safety issues identified in investigations involving Cirrus SR22T airplanes in which an excessive fuel condition led to a loss of engine power during the takeoff climb. We investigated six accidents involving Cirrus SR22T airplanes; in five of them, we were able to retrieve fuel flow data from the recoverable data modules (RDM) installed in the accident airplanes. In each case, the fuel flow data indicated an excessively high fuel flow (ranging from 42.2 to 50.1 gallons per hour [gph]) to the engine just before the loss of power.

What You Should Know

Our examination of these investigations suggests a lack of system safety assessments to identify the cause and reduce the potential of the hazard from occurring. We concluded:

the Cirrus SR22T can experience a loss of engine power due to excessive fuel flow and some causes of excessive fuel flow during takeoff and climb may not have been identified and mitigated. We have recommended that Cirrus Aircraft conduct a functional hazard assessment (FHA) to identify the causes, effects, and severity levels for the SR22T excessive fuel flow hazard condition during takeoff and climb phases of flight and, based on the FHA, update the system safety assessment. 

until the FAA requires implementation of appropriate mitigating actions to prevent the loss of engine power due to excessive fuel flow in the SR22T, additional accidents may occur due to this hazard. We have recommended the FAA review the functional hazard assessment (FHA) recommended in Safety Recommendation A-22-7 and ensure it meets the objectives of Advisory Circular 23.1309-1 E. Upon approval of the FHA, work with Cirrus to identify necessary mitigating actions and require their implementation through the appropriate means, such as an airworthiness directive. 

What You Can Do

We encourage all pilots and operators with Cirrus Aircraft to read this AIR and review the circumstances of the six accidents investigated by the NTSB.

Aviation Accident Final Report - National Transportation Safety Board

The National Transportation Safety Board traveled to the scene of this accident.

Additional Participating Entities:

Federal Aviation Administration / Flight Standards District Office; San Diego, California
Cirrus Aircraft; Duluth, Minnesota
Continental Motors; Mobile, Alabama
Hartzell Propeller; Piqua, Ohio

Investigation Docket - National Transportation Safety Board: https://dms.ntsb.gov/pubdms

http://registry.faa.gov/N707DF

Location: San Diego, CA
Accident Number: WPR18FA093
Date & Time: 02/21/2018, 0631 PST
Registration: N707DF
Aircraft: CIRRUS DESIGN CORP SR22T
Aircraft Damage: Substantial
Defining Event: Abrupt maneuver
Injuries: 1 Fatal
Flight Conducted Under: Part 91: General Aviation - Personal

Analysis 

The private pilot was taking off in the airplane when witnesses heard the airplane lose engine power at low altitude and described a sequence of events consistent with an aerodynamic stall/spin. Data retrieved from onboard the airplane recorded 47 power cycles, including the accident takeoff. The data revealed an abnormally high fuel flow, about 46 gallons per hour (gph), when takeoff power was applied, reaching 50.1 gph as the airplane entered the initial climb. About 300 ft above ground level (agl), the airplane experienced a total loss of engine power and began to decelerate and roll to the right. The airplane then entered a steep left bank and nearly vertical nose-down attitude as it descended to ground contact. The final data point at 50 ft agl indicated that the airplane was in an 81° nose-down pitch attitude and a 157° (inverted) left roll. The data is consistent with the pilot attempting to return to the runway following a loss of engine power at low altitude and an aerodynamic stall/spin when the pilot exceeded the airplane's critical angle of attack during the turn.

Examination of the engine revealed signatures consistent with an excessively rich fuel/air mixture. The fuel manifold and engine-driven fuel pump were removed from the engine and installed onto a slave engine for a test run, during which the slave engine began to surge when the power was increased to 2,300 rpm; the fuel pressure indicated 234 psi (about 39 gph); a nominal value was 210-220 psi. The engine-driven fuel pump adjustment screw was adjusted in the lean direction, and the engine operated within normal parameters; however, when the test cell fuel boost pump was turned on, the engine lost total power.

The airplane was equipped with an electric fuel boost pump that could be manually activated by the pilot via a cockpit switch. The pump had two modes; the BOOST position supplied an additional 4-6 psig at 19 gph to the engine and was used for vapor suppression when required and during takeoff, climb, landing, and when switching fuel tanks. The HIGH BOOST/PRIME position provided a capacity of 42 gph at 16 psig with a maximum full relief (no flow) pressure of 23 psig to the engine, and was used for priming the engine before start and suppressing vapor formation at flight altitudes above 18,000 ft with hot fuel.

It is likely that the excess fuel being delivered to the engine during the initial climb resulted in the total loss of power; however, the reason for the excessive fuel flow values could not be determined. According to the manufacturer, the initial fuel flow adjustments occurred at the factory about 4 months before the accident and before the pilot took delivery of the airplane. There were no maintenance logbook entries to indicate that the fuel flow had been adjusted since that time. The data did not indicate a gradual increase in fuel flow values over time; therefore, it is unlikely that the engine-driven fuel pump adjustment screw was becoming loose. It is possible that the pilot placed the fuel boost pump in the HIGH BOOST/PRIME position before takeoff, which would have increased fuel flow to about 42 gph, but this scenario does not account for the 50-gph fuel flow value reflected in the data. (The fuel boost pump position was not a recorded parameter and the position of the switch at the time of takeoff could not be determined.) Thus, based on the available information, the reason for the fuel flow rates being so high during the takeoff could not be determined.

Probable Cause and Findings

The National Transportation Safety Board determines the probable cause(s) of this accident to be:
The pilot's exceedance of the airplane's critical angle of attack during an attempted return to the runway following a total loss of engine power after takeoff, which resulted in an aerodynamic stall. Contributing to the accident was the excessive amount of fuel being delivered to the engine for reasons that could not be determined based on the available information.

Findings

Aircraft
Angle of attack - Capability exceeded (Cause)
Fuel - Fluid management (Factor)
Fuel pumps - Not specified

Personnel issues
Aircraft control - Pilot (Cause)

Not determined
Not determined - Unknown/Not determined (Factor)

Factual Information

History of Flight

Initial climb
Fuel related
Loss of engine power (total)

Approach
Abrupt maneuver (Defining event)
Aerodynamic stall/spin

Uncontrolled descent
Collision with terr/obj (non-CFIT) 

On February 21, 2018, about 0631 Pacific standard time, a Cirrus SR22T airplane, N707DF, impacted terrain shortly after takeoff from Montgomery-Gibbs Executive Airport (MYF), San Diego, California. The private pilot was fatally injured and the airplane sustained substantial damage. The airplane was registered to the pilot who was operating it as a Title 14 Code of Federal Regulations Part 91 personal flight. Visual meteorological conditions prevailed and no flight plan was filed for the flight, which was originating at the time of the accident.

Three witnesses provided statements regarding the accident. One stated that the airplane was "not very high" when the engine "went silent" and produced a small trail of black smoke that dissipated quickly. The airplane then entered a nose-down descent to ground contact. Another witness also described a sound consistent with a loss of engine power. The third witness recalled hearing the airplane's engine but stated that it was quiet. When the airplane was "just above the treeline," it turned right and shortly thereafter, "went straight down."

The airplane came to rest in a dirt construction site about 1/2 mile northwest of the departure end of MYF runway 5.

The airplane's recoverable data module (RDM) was removed for download at the National Transportation Safety Board Recorders Laboratory. The data revealed 47 power cycles. During the taxi, when the airplane turned to runway heading, the engine stabilized at a power setting of about 2,490 rpm and manifold pressure of about 35.3 inches of mercury (inHg). The fuel flow rate indicated over 46 gallons per hour (gph). As the airplane began its initial climb at 100 knots, the manifold pressure began to rise, reaching 36.6 inHg. The fuel flow also began to increase, reaching a peak value of 50.1 gph. Several seconds later, the manifold pressure decreased to its previous level of 35.3 inHg, and the fuel flow remained at 50.1 gph. Shortly thereafter, about 300 ft above ground level, the engine rpm, manifold pressure, exhaust gas temperature, and fuel flow began to decrease sharply. The airplane began to decelerate and rolled to the right. The airplane's left wing then dropped, and the airplane entered a nose-down pitch attitude. The final data point, at 50 ft agl, indicated that the airplane was in an 81°-nose-down pitch attitude with a 157° angle of bank.

The RDM captured two flights the day before the accident; review of the data revealed no anomalies. During those takeoffs, the engine stabilized at a power setting of about 2,490 rpm and 35.8 inHg manifold pressure; the fuel flow stabilized between 38 and 40 gph.

Pilot Information

Certificate: Private
Age: 61, Male
Airplane Rating(s): Single-engine Land
Seat Occupied: Left
Other Aircraft Rating(s): None
Restraint Used:
Instrument Rating(s): Airplane
Second Pilot Present: No
Instructor Rating(s): None
Toxicology Performed: Yes
Medical Certification: Class 3 With Waivers/Limitations
Last FAA Medical Exam: 08/26/2016
Occupational Pilot: No
Last Flight Review or Equivalent: 03/28/2017
Flight Time:  2700 hours (Total, all aircraft), 1000 hours (Total, this make and model), 2600 hours (Pilot In Command, all aircraft), 15 hours (Last 90 days, all aircraft), 2 hours (Last 24 hours, all aircraft) 


Aircraft and Owner/Operator Information

Aircraft Make: CIRRUS DESIGN CORP
Registration: N707DF
Model/Series: SR22T
Aircraft Category: Airplane
Year of Manufacture: 2017
Amateur Built: No
Airworthiness Certificate: Normal
Serial Number: 1621
Landing Gear Type: Tricycle
Seats: 5
Date/Type of Last Inspection: 02/16/2017, Condition
Certified Max Gross Wt.:
Time Since Last Inspection: 2 Hours
Engines: 1 Reciprocating
Airframe Total Time: 66 Hours at time of accident
Engine Manufacturer: Continental Motors
ELT: C126 installed, not activated
Engine Model/Series: TSIO-550-K
Registered Owner: On file
Rated Power: 315 hp
Operator: On file
Operating Certificate(s) Held: None

The airplane was manufactured in October 2017 and the pilot took delivery on November 1, 2017. The pilot had previously owned at least two Cirrus aircraft before the accident airplane; one was a SR22T and the other was a SR22. On November 4, 2017, the Cirrus Design Factory Service Center made a minor adjustment to the manifold pressure setting and subsequent ground engine run before the pilot's departure. According to a Cirrus Aircraft representative, the initial fuel flow adjustments occurred on October 11, 2017 at the Cirrus Aircraft factory.

At the accident site, the Hobbs meter read 66.1 hours and the flight meter indicated 52.7 hours. Review of the maintenance logs revealed a 50-hour inspection and oil change on February 16, 2018, which included an exhaust check and turbo check valve inspection.

The airplane's fuel system was equipped with an electric fuel boost pump, which could be manually activated by the pilot via a cockpit switch. The BOOST position supplied an additional 4-6 psig at 19 gph to the engine, which was used for vapor suppression when required, and during takeoff, climb, landing, and when switching fuel tanks. The HIGH BOOST/PRIME position provided a capacity of 42 gph at 16 psig with a maximum full relief (no flow) pressure of 23 psig to the engine; this position was used for priming the engine before start and for suppressing vapor formation at flight altitudes above 18,000 ft with hot fuel. Electric fuel boost pump operation was not recorded by the RDM.

Meteorological Information and Flight Plan

Conditions at Accident Site: Visual Conditions
Condition of Light: Dawn
Observation Facility, Elevation: MYF, 427 ft msl
Distance from Accident Site: 1 Nautical Miles
Observation Time: 0653 PST
Direction from Accident Site: 240°
Lowest Cloud Condition: Clear
Visibility:  10 Miles
Lowest Ceiling: Broken / 10000 ft agl
Visibility (RVR):
Wind Speed/Gusts: 6 knots /
Turbulence Type Forecast/Actual: / None
Wind Direction: 120°
Turbulence Severity Forecast/Actual: / N/A
Altimeter Setting: 30.13 inches Hg
Temperature/Dew Point: 7°C / -4°C
Precipitation and Obscuration: No Obscuration; No Precipitation
Departure Point: San Diego, CA (MYF)
Type of Flight Plan Filed: None
Destination: Yuma, AZ (YUM)
Type of Clearance: None
Departure Time: 0630 PST
Type of Airspace:

Airport Information

Airport: Montgomery-Gibbs Executive Air (MYF)
Runway Surface Type: Asphalt
Airport Elevation:427 ft 
Runway Surface Condition: Dry
Runway Used: 5
IFR Approach: None
Runway Length/Width: 3400 ft / 75 ft
VFR Approach/Landing: None


Wreckage and Impact Information

Crew Injuries: 1 Fatal
Aircraft Damage: Substantial
Passenger Injuries: N/A
Aircraft Fire: None
Ground Injuries: N/A
Aircraft Explosion: None
Total Injuries: 1 Fatal
Latitude, Longitude: 32.818889, -117.123056 (est) 

Along the northern side of the accident site was a line of undamaged trees and a light pole with right wing pieces embedded in its base. The first identified point of impact was a small crater from which two narrow indentations extended outward; one propeller blade came to rest next to the crater. Northwest of the small crater was a damaged fence and the outboard portion of the right wing. Southwest of the small crater was a damaged roadway curb with embedded white airframe fragments; other airframe fragments were noted in the immediate area. The main wreckage was located about 7 ft northeast of the initial impact point. The forward fuselage was heavily fragmented and the airframe around the cabin was destroyed. The left wing trailing edge was embedded in the dirt and the wing skin was fracture separated. The inboard right wing was still mostly secured to the airframe; however, the outboard portion was fracture separated. The aft fuselage and empennage were mostly intact and undamaged. The Cirrus Airframe Parachute System (CAPS) parachute was found extended from the aft fuselage; however, it was unopened.

Postaccident examination of the airframe revealed that the CAPS activation handle was stowed and the safety pin remained installed. Flight control continuity was established throughout the airframe. The flaps were in the retracted position. All three propeller blades sustained damage to the trailing edges; two blades were still attached to the propeller hub.

Postaccident examination of the engine revealed operating signatures consistent with an overly rich fuel/air mixture. The spark plugs were sooty and black in color. The cylinders were removed from the engine; the combustion chambers and valve heads were very sooty. In addition, the piston heads exhibited combustion deposits and were sooty. Examination of the turbocharger system revealed spiral streaking on the turbine housing outlet. The fuel manifold and engine-driven fuel pump were removed from the engine and installed onto a slave engine for an engine run.

The slave engine was installed onto a test bench and started normally. After idling for a short time, the power was increased to 2,300 rpm and the engine started to surge; the fuel pressure indicated 234 psi (about 39 gph) and the engine was unable to produce full power. The fuel pressure specification for the engine is 210-220 psi. The engine-driven fuel pump adjustment screw was measured and adjusted in the lean direction. The engine was restarted, and the engine operated up to 2,566 rpm (full power) with a fuel pressure reading between 200 and 213 psi. The test cell fuel boost pump was turned on and the engine lost all power. The engine-driven fuel pump was adjusted back to its original position and the engine was restarted. The throttle was increased to 2,404 rpm and the engine began to surge again. The fuel pressure gauge varied from 220 psi to 261 psi (about 44 gph). Toward the top of the surge, the fuel pressure gauge stayed between 250-260 psi when it would spike at about 260 psi, then return to about 220 psi.

Further examination of the fuel pump revealed that a green paint stripe was present on the adjustment screw and nut; the paint stripe had been previously broken and the paint on the nut did not match the paint on the screw (before adjusting it in the test cell). The green paint is consistent with being applied when the pump was assembled and tested. According to Cirrus Aircraft, once the engine is assembled, it is then shipped and installed onto the airplane, where Cirrus Aircraft employees make changes to the adjustment screw during the engine set up and precertification flight test. Cirrus does not apply torque stripe or paint after making these adjustments, nor are they required to. 


Medical And Pathological Information

An autopsy of the pilot was performed by the San Diego County Coroner, San Diego, California. The cause of death was listed as multiple blunt force injuries.

Toxicology testing was performed by the Federal Aviation Administration Forensic Sciences Laboratory with negative results for carbon monoxide, ethanol, and tested-for drugs.

Dr. John Harvey Serocki
Dr. Serocki graduated from the University of California, San Diego with a degree in applied mechanics and engineering sciences and went on to earn a master's degree in mechanical engineering from the Massachusetts Institute of Technology. He subsequently earned a medical degree from Northwestern University, Feinberg School of Medicine. He practiced as an orthopedic surgeon for more than 25 years with a specialization in Sports Medicine and Hand Surgery. 







10 comments:

  1. Any flyers out of Montgomery-Gibbs , was this crash on airport?

    ReplyDelete
    Replies
    1. This is in the post
      “The airplane came to rest in a dirt construction site about 1/2 mile northwest of the departure end of MYF runway 5“

      Delete
  2. 1.0 Scene Overview

    The airplane came to rest on the corner of a construction site about one mile northeast of the airport center, which is about 1/2 mile from the departure end of runway 5.

    The accident site consisted of level terrain with with loose dirt and rocks. Prior to the accident, the construction crews had dug a trench about 2 feet wide and 3 feet deep. The trench had a curb (from a previous road) stretching diagonally across it.

    There were several trees and a light pole along the northern edge of the accident site; the trees were undamaged.

    The light pole was leaning slightly to the west with fragments of right wing were wedged into the base of the pole, otherwise it was undamaged.

    The outboard portion of the right wing was separated into two pieces that came to rest at the base of the pole.

    ReplyDelete
  3. I thought Cirrus was a plane marketed as fully automated. Just put gas in and go. No need to manage a thing just engage the autopilot and fly to your destination. Get into trouble? Pull the big red handle and float down safely under a chute.
    My "opinion" is that engine gave plenty of warning that the mixture was excessively rich, maybe during taxi or run up and very likely on the takeoff roll. Abort point!
    Turns out there is still some skill level to piloting one of these contraptions after all. Just because it's a newer bird doesn't mean we can be lax. Sad outcome.

    ReplyDelete
    Replies
    1. I take it you are a non-Cirrus owner. Cirrus SR22T takes HIGH skill to pilot. Fuel and turbo management is complex and requires both knowledge and training. In fact, a Cirrus SR22T is the top of the food chain in GA piston cross-country aircraft.

      Delete
  4. Previous flight.
    NYL
    YUMA, AZ
    MYF
    SAN DIEGO, CA
    TUESDAY 20-FEB-2018
    12:53 MST
    TUESDAY 20-FEB-2018
    12:45 PST
    Tue 14:53:53 Departure (KNYL) @ Tuesday 12:53:53 MST FlightAware ADS-B (KNYL)
    Tue 14:53:53 32.6810 -114.6148 ← 310° 110 127 800 Level FlightAware ADS-B (KNYL)
    Tue 14:54:23 32.6905 -114.6312 ← 303° 124 143 1,300 1,100 Climbing FlightAware ADS-B (KNYL)
    Tue 14:54:53 32.6997 -114.6483 ← 305° 126 145 1,900 1,400 Climbing FlightAware ADS-B (KNYL)
    Tue 14:55:23 32.7093 -114.6652 ← 306° 120 138 2,700 1,560 Climbing FlightAware ADS-B (KNYL)
    Tue 14:55:43 32.7158 -114.6761 ← 306° 119 137 3,200 1,412 Climbing FlightAware ADS-B (KNYL)
    Tue 14:56:14 32.7251 -114.6923 ← 306° 125 144 3,900 1,377 Climbing FlightAware ADS-B (KNYL)
    Tue 14:56:44 32.7350 -114.7090 ← 305° 120 138 4,600 1,263 Climbing FlightAware ADS-B
    Tue 14:57:11 32.7429 -114.7221 ← 304° 123 142 5,100 1,418 Climbing FlightAware ADS-B (KBXK)
    Tue 14:57:39 32.7458 -114.7430 ← 271° 130 150 5,900 1,472 Climbing FlightAware ADS-B (KNYL)
    Tue 14:58:04 32.7460 -114.7612 ← 269° 134 154 6,400 1,364 Climbing FlightAware ADS-B (KNYL)
    Tue 14:58:23 32.7458 -114.7748 ← 269° 135 155 6,900 1,200 Climbing FlightAware ADS-B (KBXK)
    Tue 14:59:14 32.7469 -114.8132 ← 275° 155 178 7,800 806 Climbing FlightAware ADS-B (KBLI)
    Tue 14:59:30 32.7475 -114.8283 ← 273° 172 198 7,800 98 Climbing FlightAware ADS-B (KBLI)
    Tue 15:00:15 32.7481 -114.8723 ← 272° 186 214 7,900 320 Climbing FlightAware ADS-B (KBLI)
    Tue 15:00:45 32.7485 -114.9032 ← 272° 184 212 8,200 490 Climbing FlightAware ADS-B
    Tue 15:01:04 32.7489 -114.9236 ← 272° 184 212 8,300 120 Climbing FlightAware ADS-B (KCXL)

    ReplyDelete
  5. Youth error... every mechanical system has a bathtub reliability curve i.e youth error-maturity-old age.

    So the first 100 hrs of a limited production expensive piece of machinery need to be under scrutiny as a variety of new systems being broken into including the engine are settling down in their tolerances.

    Looks from the NTSB docket the fuel pump was set too high and was flooding the engine reducing its power significantly. The trail of smoke is also a clue of a too rich mixture.

    ReplyDelete
  6. This accident looks similar to Mooney M20V N576CM. Brand new plane goes down with engine failure. Just guessing.

    ReplyDelete
  7. The exhaust gas temperature i.e. TIT would have been reduced due to the rich mixture. Was this not measured or tracked in this aircraft?

    ReplyDelete
    Replies
    1. EGT is Exhaust Gas Temp. TIT is Turbo Inlet Temp. A rich mixture would have caused TIT temps to drop in cruise. EGT may have been affected in the climb...but neither of them would have given a warning of an overly rich mixture. Usually the plane sputters and coughs and you can FEEL the power drop when there is to much fuel. This engine power loss due to excess fuel may have been unavoidable if the fuel pump was not set correctly.

      Delete