N310XXCessna 310 · 1956

On October 3, 2010, about 1535 Pacific daylight time, a Cessna 310, N310XX, was substantially damaged when it impacted terrain shortly after takeoff from runway 22 at Catalina Airport (AVX), Avalon, California. The non-instrument rated owner-pilot and one passenger received minor injuries, and one passenger received serious injuries. The personal flight was operated under the provisions of Title 14 Code of Federal Regulations (CFR) Part 91. Meteorological conditions were changing rapidly, and no flight plan was filed for the flight.

According to the pilot, he and the passengers landed at AVX on Catalina Island the previous day, and the passengers remained on the island while the pilot continued on to John Wayne-Orange County Airport (SNA), Santa Ana, California. On the day of the accident, the pilot departed SNA a few hours late due to weather, and landed at AVX, where he and the passengers then began lunch. While dining, the pilot noticed that the weather was deteriorating, and suggested that they depart before instrument meteorological conditions prevailed. The passengers and pilot boarded the airplane, and the pilot started the engines. Due to the deteriorating weather and the fact that he had just flown in from SNA, the pilot conducted what he termed an "abbreviated" engine run-up during the taxi-out. The pilot planned to conduct what he referred to as a "Vx takeoff," which entailed an initial climb at the airplane's best angle-of-climb speed. According to the pilot, the best angle-of-climb speed (Vx) was approximately the same as the minimum controllable airspeed (VMC), which was about 85 mph.

The pilot stated that the takeoff roll was normal, but about 2 to 3 seconds after liftoff, the airplane veered "sharply to the left," which he interpreted as a failure of the left engine. Because he was concerned about airspeed decay due to the flaps and landing gear being extended, the pilot pushed the nose down to maintain airspeed. He then noticed the "right wing coming up," so he retarded the right throttle. The airplane then entered a cloud/fog bank, and impacted terrain. It came to a stop quickly, with the cabin intact, but very shortly thereafter was engulfed by fire. All three occupants exited the airplane without external assistance.

The wreckage was examined on-site by National Transportation Safety Board (NTSB), Federal Aviation Administration (FAA), and manufacturers’ personnel. It was recovered to a secure facility for detailed examination and testing. Subsequent to that, the engines were removed and sent to Teledyne Continental Motors in Mobile, Alabama, and the left fuel selector valve was sent to Precision Airmotive in Marysville, Washington, for additional testing

According to FAA information, the 54-year-old pilot held a private pilot certificate with airplane single- and multi-engine land ratings. His most recent FAA third-class medical certificate was issued in December 2009. According to the pilot, he had approximately 700 hours of total flight experience, of which 650 hours were in the accident airplane. He reported that his personal flight records were on board the airplane at the time of the accident, and were consumed in the fire. An email from his flight instructor stated that the pilot's most recent flight review was completed on July 12, 2010. Despite multiple requests, the pilot did not provide a completed accident reporting form to the NTSB.

General Information

The airplane was manufactured in 1956, and was registered to the pilot in 2004. It was equipped with two Teledyne Continental Motors (TCM) O-470-M engines, each driving a controllable pitch, metal, two-bladed propeller. The airplane was equipped with tricycle-configuration retractable landing gear, and split-style trailing edge flaps.

Fuel System

The airplane was equipped with two 50-gallon tip tanks for fuel. Each tank contained a cockpit-controlled boost pump. The airplane was equipped with two fuel selector valves, one for each engine. A fuel strainer was located downstream of each fuel selector valve. Each engine was equipped with an engine-driven fuel pump and a pressure carburetor.

Two separate rotary-style fuel selector controls, one for each engine, were mounted side-by-side in a recess on the cockpit floor between the two front seats. Each valve control layout was the same, with three possible setting positions; one each at the 9 o'clock (left), 6 o'clock (aft), and 3 o'clock (right) position. Those positions were respectively labeled "LEFT ON TANK," "BOTH OFF," and "RIGHT ON TANK." Setting the left selector control to "LEFT ON TANK" would provide fuel to the left engine from the left tank, while setting it to the "RIGHT ON TANK" would enable a cross-feed setting, where the left engine would feed from the right tank. The right selector valve controls and plumbing were similarly configured.

Each selector valve was mounted on the outboard aft wall of its respective engine nacelle. A series of mechanically linked rods, oriented transversely with respect to the airplane, coupled each selector valve control to its respective valve. Actuation of the left or right cockpit fuel selector control rotated the respective rod assembly about its longitudinal axis, which rotated the internal valve mechanism, enabling selection of the tank to be used to provide fuel to each engine. There was a one-to-one ratio between the rotation of the cockpit control and the rotation of the selector valve; rotating the control 90 degrees rotated the valve 90 degrees. Detents for each of the selected positions, each 90 degrees apart, were located on the valve body.

The manufacturer's recommended engine start and takeoff procedure was to feed each engine from its respective fuel tank. In that configuration, the left fuel selector control would point to the left, and the right selector control would point to the right. Each selector valve control would point aft when "BOTH OFF" was selected.

Maintenance Records Information

According to the maintenance records, the most recent annual inspection was completed in June 2010. The records indicated that at that time, the airplane had accumulated a total time in service (TT) of 5,520 hours. The engine records indicated that as of that annual inspection, both the left and right engines had accumulated a TT since major overhaul (TSMOH) of 1,713 hours. The airplane and engines had accumulated about 500 hours in the 5 years prior to the accident. Examination of the records for the previous 5 years did not reveal any noteworthy relevant mechanical issues.

The 1451 automated weather observation at AVX included winds from 280 degrees at 7 knots; visibility 10 miles, clear skies; temperature 19 degrees C; dew point 14 degrees C; and an altimeter setting of 30.00 inches of mercury.

The 1532 observation included winds from 240 degrees at 11 knots; visibility 2 miles in haze, few clouds at 100 feet; temperature 17 degrees C; dew point 14 degrees C; and an altimeter setting of 29.99 inches of mercury.

The 1534 observation included the same values as the observation 2 minutes prior, except for visibility 1 mile in mist, a broken cloud layer of indeterminate height, and a vertical visibility of 200 feet.

According to FAA Airport/Facilities Directory information, AVX was equipped with a single runway, designated 04/22, which was paved, and measured 3,000 feet by 75 feet. Airport elevation was 1,602 feet above mean sea level (msl). The airport was not equipped with an air traffic control tower.

The airplane came to rest upright, about 880 feet southeast of the runway, at an elevation about 80 feet below the runway elevation. The terrain of the impact area was flat and sloped down in the direction of travel, with some small cactus and scrub vegetation. Ground scars indicated that the airplane first impacted the terrain about 660 feet from the runway, and about 60 feet below the airport elevation. The ground scars were aligned on a magnetic heading of approximately 150 degrees, and the airplane came to rest on a magnetic heading of approximately 072 degrees. The fuselage and cabin exhibited severe fire damage. Most of the cabin shell, and most of the interior components, exclusive of steel items, were consumed by fire.

Exclusive of the fire damage to the fuselage, the airplane was essentially intact, and all major components, with the exception of the tip tanks and landing gear, were attached and in their normal relative positions. The tip tanks were separated from the airplane and substantially damaged, so no fuel quantity information could be obtained. The empennage and wings, including the nacelles and engines, exhibited minor thermal damage. All flight control surfaces were present and attached to their aerodynamic surfaces, and control continuity was established. The setting positions for the cockpit flight and engine controls and switches could not be determined.

Both engine cowlings were impact damaged, but both engines were in good condition, with little or no thermal damage. Engine impact damage was primarily limited to the exhaust piping and some other hardware on the undersides of both engines. The propeller blades remained attached to their respective hubs, and the hubs remained attached to their respective engines. No blades appeared feathered, and all blades were bent aft, with minimal leading edge and rotational signature damage. On scene examination of the engines and propellers did not reveal any pre-accident conditions or failures that would have precluded normal operation.

The fuel selector valve for the right engine was found set to the right fuel tank. The fuel selector valve for the left engine was found displaced approximately 30 degrees from the off position, towards the left tank setting.

The engines were removed and shipped to the Teledyne Continental Motors in Mobile, Alabama, for additional examination and testing. The left fuel selector valve was removed for additional testing.

Refer to the following section and the accident docket for detailed information.

Owner's Manual and Pilot-Produced Checklists

According to the pilot, he utilized personal, customized checklists for his operation of the airplane. He stated that the checklists were derived from the manufacturer's owner's manual (OM) and modified to a "more efficient and comprehensive format." Several companies that do not manufacture aircraft produce and sell such modified checklists for aircraft owners and operators. Some aftermarket or more recent manufacturer's checklists incorporate double checks, in the form of subsequent duplicate entries, for items, such as fuel selector position, that are critical to flight safety.

The pilot's checklist used on the accident flight was consumed in the fire, but he provided a document, which he said was the same as the one he used. He stated that his checklist was contained on two sides of a single sheet of paper. Comparison of the pilot-provided checklist with that in the OM revealed that the OM tended towards phrases or sentences for each item, whereas, the pilot's checklist items were primarily single-word "challenge and response" style entries. The OM checklist utilized bold font for the section/phase of flight headers, and numbered the steps in each section, while the pilot's checklist, including section/phase of flight headers, was unnumbered, virtually all a single font style, and cluttered in appearance. In addition, multiple hand-written entries were interspersed among the printed entries on the pilot's checklist. The OM checklist contained a "Before Starting Engine," a "Starting Engine," a "Warm-Up And Ground Test," and a "Before Take-Off Or During Taxi" section. In contrast, the pilot's checklist appeared to combine those sections into two sections entitled "Start-Up" and "Run-Up."

Step 9 of the OM "Before Starting Engine” checklist appeared as follows:

(9) Check left engine fuel selector valve "ON LEFT TANK," and right engine fuel selector valve “ON RIGHT TANK.”

In contrast, the pilot’s corresponding personal checklist entry was:

Neither the OM checklist nor the pilot’s personal checklist subsequently re-addressed the fuel selector valve position in any of their respective sections for start, runup, or takeoff.

Shutdown and Pre-Takeoff Procedures

In a post accident interview with the NTSB, the pilot stated that when securing the airplane for every shutdown, he would run the engines "lean" for a few minutes at an rpm of about 1,200. He would then shut down the engines via the mixture control, and after that, he would turn the fuel selector valves to "off." He stated that he followed those procedures after landing at AVX on the day of the accident. The checklists provided by the pilot did not contain any post-landing or engine shutdown procedures. Examination of the OM "After Landing" checklist revealed that it did not contain any steps regarding the fuel selector valves.

During that same interview, as well as in another interview with FAA personnel, the pilot stated that he conducted an "abbreviated" engine run-up prior to the accident departure, since he had conducted a full and satisfactory run-up prior to leaving SNA. He explained that the abbreviated run-up consisted of a magneto check during taxi-out.

In a subsequent email communication to the NTSB, the pilot modified or amplified his account of some of the information he had previously provided. Regarding his previous statement that he turned the fuel selectors off after every flight, the pilot stated that that was "not procedural" to him, and that sometimes he did not shut the fuel off. Regarding his previous description of the accident flight engine run-up, he amended the description by adding "the only item missing on the run-up would have been prop feather." In his original recount of the event, he stated that he advanced to engines to 2,200 rpm prior to brake release. His revised account stated that he ran the engines to 2,200 rpm for 5 to 10 seconds while the airplane was on the concrete apron adjoining the threshold of runway 22 at AVX.

Single Engine Operating Airspeeds

Chapter 12 ("Transition to Multiengine Airplanes") of the FAA Airplane Flying Handbook (AFH, FAA-H-8083) contained the following information:

"The basic difference between operating a multiengine airplane and a single-engine airplane is the potential problem involving an engine failure. The penalties for loss of an engine are twofold: performance and control. The most obvious problem is the loss of 50 percent of power, which [significantly] reduces climb performance…. The other is the control problem caused by the remaining thrust, which is now asymmetrical. Attention to both these factors is crucial to safe one engine inoperative (OEI) flight."

The AFH further noted that:

"Twin-engine airplanes have several additional performance "V" speeds unique to OEI operation. These speeds are differentiated by the notation "SE," for single engine."

Excerpted key AFH definitions or descriptions included:

VXSE - Best angle-of-climb speed with one engine inoperative.

VYSE- Best rate-of-climb speed with one engine inoperative. Marked with a blue radial line on most airspeed indicators. Above the single-engine absolute ceiling, VYSE yields the minimum rate of sink.

VSSE– Safe, intentional one-engine-inoperative speed. Originally known as safe single-engine speed. Now formally defined in Title 14 of the Code of Federal Regulations (14 CFR) Part 23, Airworthiness Standards, and required to be established and published in the AFM/POH. It is the minimum speed to intentionally render the critical engine inoperative.

VMC – Minimum control speed with the critical engine inoperative. Marked with a red radial line on most airspeed indicators. The minimum speed at which directional control can be maintained under a very specific set of circumstances outlined in 14 CFR Part 23, Airworthiness Standards. There is no requirement in this determination that the airplane be capable of climbing at this airspeed. VMC only addresses directional control.

Accident Airplane Certification Basis and Single Engine Speeds

According to the FAA Type Certificate Data Sheet (TCDS), the certification basis for Cessna Model 310 serial number 35411 was Civil Air Regulation (CAR) 3, 3 dated November 1, 1949, as amended by 3-1 through 3-10. The only reference in the TCDS to single engine control speed was the statement that required the airplane to be equipped with a placard on the instrument panel that stated "Minimum speed for single engine operation 95 mph. (TIAS)."

The certification basis did not require either the red radial line denoting VMC, or the blue radial line denoting VYSE, to appear on the airspeed indicator (ASI). Those markings were only mandated for airplanes certificated under Part 23, which became effective about 1964. Neither the TCDS nor the OM explicitly specified or discussed any colored arcs or radial lines on the ASI. The investigation was unable to determine whether the pilot’s airspeed indicator was marked with either the VMC radial red line or the VYSE radial blue line.

In addition, neither the FAA nor the airplane manufacturer mandated or recommended such VMC or VYSE markings on the ASIs of the accident airplane make and model.

Single-Engine Airspeed Guidance Available to the Pilot

The "Operating Limitations" section of the manufacturer's OM did not include any references to VMC, VSSE, VYSE, or any other single engine operating speeds. A review of the OM revealed that it included only one explicit reference to "minimum single engine control speed," which appeared as step (6) in the "Normal Take-Off" checklist in the "Operating Checklist" section. Step (6) stated "After take-off, level off and accelerate to 93 mph (minimum single engine control speed)." Fire damage precluded verification of whether the airplane was equipped with the placard.

The OM contained two checklists for engine failure during takeoff. The first was entitled "ENGINE FAILURE DURING TAKE-OFF BELOW 93 MPH;" the only checklist step was "Cut power and decelerate to a stop." The second checklist was entitled "ENGINE FAILURE DURING TAKE-OFF ABOVE 93 MPH WITH OBSTRUCTIONS AHEAD." In sequence, the checklist included steps to increase the good engine to maximum power, retract the landing gear, identify the inoperative engine, and feather that propeller. The checklist then provided target airspeeds as a function of flap setting. The copy of the pilot’s personal checklist did not contain any reference to minimum control airspeed or the value itself (93 mph). The pilot’s personal checklist did contain a section entitled "SINGLE ENG CLIMB," which included the challenge-and-response style entry "ASI 109 min BLUE +" which was a reference to the blue line on the ASI.

In his communications to the NTSB, depending on his account, the pilot indicated that the minimum control airspeed was 85 to 90 mph. Both values are below the published value of 95 mph, and the "109" value in the pilot’s personal checklist was well above that value. Fire damage precluded verification of whether the airplane was equipped with the OM, or whether the ASI was marked with the blue VYSE line, even though it was not required. The actual minimum control target airspeed used by the pilot during the accident flight could not be determined.

Pilot’s Post-Engine Failure Decision-Making

In his communications to the NTSB, the pilot indicated that he frequently practiced the procedures for engine failure on takeoff, and that he was quite familiar with the limited airplane climb capability with the landing gear and flaps extended. He also stated that he was cognizant of the importance of maintaining the minimum safe airspeed, in order to prevent a loss of aerodynamic control. When the engine failure occurred, the airplane yawed sharply to the left. The pilot was confronted with the choice of attempting to continue the takeoff and climb over the cloud/fog bank, or pushing the nose down in order to maintain the minimum safe airspeed. The pilot stated that given his knowledge of the airplane’s limited single engine climb capability in its current configuration (gear and flaps extended), he opted to abandon the takeoff, and pitched down into the cloud/fog bank, which he described as a "whiteout." At that point, he retarded the right throttle, with the intent of making a forced landing on the unseen terrain ahead. In his post accident discussions with the NTSB, the pilot stated that he never attempted to raise the landing gear or flaps, nor did he attempt to feather the left propeller.

The investigation was unable to locate any data to enable the determination of the trajectory or speed of the airplane at any point during the flight. However, the evidence indicated that the airplane struck the terrain in a relatively wings-level attitude, which was consistent with the pilot maintaining aerodynamic control of the airplane.

Engine Examinations and Test Runs

Examinations and test runs of the two engines were conducted at the facilities of Teledyne Continental Motors, Mobile, Alabama. Both engines exhibited characteristics and appearances consistent with their age, and both sustained impact damage. The top spark plugs were removed from the left engine; they were sooty but otherwise normal. One top plug from the right engine was oily, but the rest were normal. Each engine was able to be freely manually rotated. Crankshaft and valve train continuity, accessory drive integrity, magneto impulse coupling activation, and thumb compressions were obtained or confirmed on both engines.

All 12 cylinders were borescoped, and all cylinder head combustion chambers, intake and exhaust valve faces, piston heads, and cylinder bores exhibited normal operating signatures. Oil was present on all cylinder bores. The left-engine cylinder bore finishes were steel, while the right-engine cylinder bore finishes were cermi-nil. No evidence of preimpact mechanical malfunction was noted during the examination of either engine.

Pre engine-run cylinder compressions were accomplished and recorded, and all impact-damaged components were replaced on both engines. Separately, a test propeller was installed on each engine, each engine was installed in a test cell, and test run. Both engines started readily and performed normally throughout their normal operating ranges. Both engines accelerated normally without any hesitation, stumbling or interruption in power, and demonstrated the ability to produce rated horsepower. The operation of each engine was normal, and the tests did not reveal any abnormalities that would have precluded normal operation or production of rated horsepower.

Refer to the accident docket for detailed information.

Left Engine Fuel Selector Valve

The selector valve for the left engine was removed from the airplane, and subjected to functional evaluation and subsequent flow testing. Low pressure air was used to determine which passageways/ports were opened or closed as a function of the position/orientation of the 'flat' on the actuation rod fitting. That evaluation determined that the valve operated per design; the fitting orientation corresponded to the proper opening or closing of the appropriate ports on the valve.

Since the left engine was successfully run at full power in a test cell after the accident, and the left selector valve was found on-scene set to a position between off and the appropriate fuel tank, a test was conducted to determine valve flow capability as a function of actuator fitting (and therefore valve setting) position. Calculations based on engine performance chart and ambient condition information indicated that at its takeoff rpm setting, the engine fuel flow would be approximately 21 to 23 gallons per hour (gph). Nominal input pressure from the fuel tank (with boost pump operating) was determined to be approximately 9 to 11 pounds per square inch (psi). Values of 5, 10, and 15 psi were used for the input pressure test matrix. For safety considerations, the test fluid was not avgas, but it had a similar specific gravity.

Tests with the valve in its normal "on" position indicated that at input pressures of 5, 10 and 15 psi, the valve would permit flow rates of 36, 58, and 75 gph, respectively. When the valve was set halfway between off and on, those values decreased to 32, 50, and 65 gph, respectively. When the valve was set to the as-found position, the respective flow rates were 0.3, 0.4 and 0.05 gph. Those values were significantly below the required fuel flow rate for the engine at takeoff power.

Refer to the accident docket for additional details.

Other Cessna 310 Takeoff Accident

Several months after his accident, the pilot of N310XX became aware of a fatal accident involving another Cessna 310. The pilot of N310XX told the NTSB that he understood that that other accident was caused by interference between the alternate landing gear handle and a fuel selector valve, which caused one of the engines to fail shortly after takeoff. The N310XX pilot further asserted that the FAA had mandated design changes as a result of that previous accident. The pilot did not provide any substantiating data regarding either the specifics of the problem, or of the alleged FAA corrective action, but he suggested that that same scenario was a possible cause for his accident.

Research revealed that the pilot of N310XX was referring to NTSB accident number LAX02FA214, which occurred on July 4, 2002, in San Dimas, California. In February 2003, the FAA issued Special Airworthiness Information Bulletin (SAIB) CE-03-23 that stated that if the alternate gear "handle is left unseated, there is the potential that operation of the primary gear motor may rotate the handle causing it to extend and strike the left engine fuel selector, and drive the selector to the off position." Additional information from Cessna and the FAA clarified that the handle would have to be positioned between its stowed and operating positions, and that the landing gear would have to be in transit. There was no evidence that either of these conditions existed on the accident flight, and the fact that the landing gear remained locked in its extended position is a contraindication to the pilot's postulated scenario.