Eyewitness Account
After 33 years of accident-free flying, I experiencedthe worst horror I have yet seen around 1:30 PST this Thanksgiving Day. I had flown theTwin Comanche from Mariposa to Carson City, a trip I have made many times over the years,to pick up my brother for the Thanksgiving holiday. We grabbed a quick bite to eat in townafter I arrived at O04 and returned to the airport and headed promptly for the Comancheparked in front of the old Carson City terminal building. As we were walking toward theplane, I noticed an Aztec departing runway 27 and I was somewhat surprised to see italready several hundred feet in the air as it climbed to the west only a few hundred feetin front of me. Having learned to fly at Carson City in the early 60s and havingflown twins out of the same airport for many years, even on a cool day with slightly lessthe standard conditions, the density altitude was still very close to 4000 feet. TheAztec, in short, had more altitude than I usually see for a twin departing Carson City.
What made me look on, I don’t know. But right before my eyes the most tragic Vmcdemonstration I could ever have imagined unfolded in a matter of no more than four or fiveseconds. I would estimate the Aztec was close to 300 feet AGL as it passed my brother andme about midfield. Suddenly the synchronous sound of the engines changed and the planepitched up slightly. It became obvious that an engine had lost power and a lot of it. Iwatched in horror as the right wing began to rise and shouted to my brother, "Oh, myGod." The Aztec continued in its roll to the left, went inverted, and then atthat point the tail pointed vertical towards the sky as the nose aimed at the ground andthe uncontrolled plane continued its leftward spin towards the ground. It disappearedbehind the row of hangars southwest of the field and perpendicular to the runway. Althoughthe actual impact was not visible to me, it appeared the Aztec had made about one totalturn in the spin and was inverted on its path to terra firma. A flat thud came from thehangar area as smoke began to rise from behind the hangars.
I have seen other accidents ata farther distance before, and I knew this one was not minor. Several people near thehangars began running toward the crash site, so I ran to a pay phone only 50 feet awayto call 911. But before I could get the first digits dialed, I could hear the firedepartment on the field responding. My brother and I ran past the hangars to be stopped bya deputy sheriff who sped onto the scene. We couldn’t see much from where we were atthat point. The plane had struck the second row of hangars and a huge cloud of black smokewas bellowing skyward as the fire crew dragged their hose toward a mangled charcoalhangar. I gave my business card to the deputy, told him with regret that I was acorporate pilot, multiengine instructor, and aircraft mechanic, and that I hadunfortunately been closer than anyone else in seeing the entire drama unfold. I told himto have the NTSB contact me if they needed a witness. After waiting for about an hour asthe emergency crews suppressed the fire and removed the victims, my brother and I flewback to Mariposa since we didn’t want to be in the way.
I have never been more alerton takeoff in the Twin Comanche in my life. I also had a pit in my stomach that madeThanksgiving dinner anything but digestible, especially since the rumor had circulatedthat there were probably four people in the plane. I doubt that any survived.
Much attention will be given to why the engine failed in theinvestigation. However, what was so vividly clear to me was the inappropriate handling ofthe engine failure. I could see no smoke coming from the engine. I may have seen somethingfly off of the engine. Because the good engine was drowning out the anemic engine, it washard to tell what kind of failure it was. I think I heard a couple of pops. What Ican’t understand is why the pilot at first pitched up. The plane appeared to me to beunusually high and slow (below blue line at least) before the event occurred. Instead ofpitching forward, and instead of pulling power on the right (the good engine), the pilotapparently panicked and tried to make the plane stay in the air when it was already wellbehind the envelope. If I had had a video of this accident, I would make multiple copiesand spend some serious time analyzing pilot actions with my multiengine students when apowerplant fails immediately after takeoff. I sincerely believe the Aztec may have beenable to land on the remaining available runway if swift and positive actions had takenplace in the correct manner. At the worst the Aztec would have landed under controloutside the airport boundary with the occupants having some chance of survival. Witnessingthis terrible accident on this Thanksgiving day is something I wish had never been. Butevery fleeting second, every detail of the roll inverted, every futile command I shoutedto a pilot who could not hear me to get the nose down and pull the power back, and thedesperate hope that somehow the plane would recover the very last second will remainindelible in my mind forever.
Is it possible to give pilots training in real life situations such as this one withoutjeopardizing safety? Can the simulator gurus come up with an inexpensive, virtual realitytraining package that can take us pilots through these critical emergencies so we wouldhandle them as perfunctorily as shaving our face with a razor every day?
My heart goes out to the pilot and the occupants of the Aztec. I wish I were a magiciancapable of recreating the entire drama with a successful outcome. As it is, I learned morein four or five seconds about an engine failure after take off, than I have after yearsand years of flying. But there has to be a less expensive way.
— Eric Gourley
Probable Cause Report by the
National Transportation Safety Board
Accident occurred NOV-27-97 at CARSON CITY, NV
Aircraft: Piper PA-23-250, registration: N6933Y
Injuries: 4 Fatal.
NTSB Identification: LAX98FA048
Witnesses reported that the aircraft’s engines sounded normal as the aircraft climbedat an unusually steep angle after takeoff to about 300 feet agl, then stalled, rolled tothe left, and spun into a hangar. One witness reported that as the aircraft rolled over hecould see that the nose baggage door on the right side of the nose was open and foldedback in front of the windshield as the aircraft continued to roll. The pilot had 487 hourstotal flying time with 23 hours in multiengine aircraft. He had received an 8.6 hourcheckout in the accident aircraft 2 weeks prior to the accident.
History Of Flight
On November 27, 1997, at 1235 hours Pacific standard time, a Piper PA-23-250, N6933Y,rolled inverted during the takeoff initial climb and impacted a hangar at the Carson City,Nevada, airport. The aircraft was destroyed and the commercial pilot and three passengerswere fatally injured. Visual meteorological conditions prevailed for the personal flight.The flight was originating at the time and was destined for Nacadoches, Texas.
A private pilot witnessed the takeoff from near the midfield location adjacent torunway 27. The runway is 5,900 feet long and the airport elevation is 4,697 feet. Thewitness reported that the accident aircraft lifted off prior to reaching his location andpassed in front of him at 30 – 50 feet agl. The nose gear had retracted and the main gearwas in transit. The engines sounded normal. As the aircraft continued west over therunway, the witness sensed that something was wrong as the aircraft reached 150 – 200 feetagl, it yawed slightly, and appeared to mush. The nose pitched up and the aircraft rolledto the left, then the nose dropped and the aircraft disappeared behind a parked aircraftin a vertical dive and he heard the impact in the hangar. When the aircraft pitched up androlled to the left, he had a view of the top of the aircraft from the rear and could seethat the nose baggage door on the right side of the nose was open. As the aircraftcontinued to roll, the baggage door appeared to fold back over the nose in front of thewindshield but it did not come off. He did not observe anything fall from the aircraft andthe engine sound was normal except that there was what the witness termed a "Dopplershift" as the aircraft spun toward the ground.
Another witness, a commercial pilot who was on the midfield ramp about 200 feet southof the runway edge, also reported that the engines sounded normal as the aircraft passedmidfield. His attention was attracted to the aircraft by its abnormally steep climb angle.However, this witness reported hearing a pop – pop sound which he attributed to one of theaircraft’s engines immediately before the nose pitch up and "VMC roll".
Another pilot/witness observed the aircraft climbing out at a very steep angle andstated, "As we continued to watch the plane continued up on a steep angle, until theplane stalled. At the stall the plane nosed down to the left and went directly down into a[hangar]."
Still another witness reported that the aircraft appeared to climb normally to about200 feet agl and then it "appeared to zoom upward at an excessive rate for about 100ft. more and then nosed over in a dive straight for the ground."
Another commercial pilot, who did not witness the accident, reported landing at CarsonCity about 1130 behind the landing Aztec and taxiing behind the aircraft to the(unattended) self-fueling pumps. He said that there were two young men in the aircraft andhe spoke to the one who was the pilot. The pilot said that he had rented the aircraft inSouthern California and had flown to Santa Rosa to pick up his brother. They were nowpicking up his mother and girlfriend and going to Texas. While they were talking a cardrove up with the two women. The other young man, who did not appear knowledgeable aboutairplanes, asked the pilot where to load the baggage. The pilot said to put the heavy bagsup front and the second man loaded approximately two to four pieces of luggage in the nosebaggage compartment. It appeared that the pilot fueled all the tanks of the accidentaircraft to capacity. This pilot fueled his aircraft after the pilot of the accidentaircraft fueled his and recalled clearing the pump which read about 109 gallons.(According to the facility operator, there is only one grade of fuel sold at theself-fueling facility, 100LL.) This pilot then took off about 1200 while the accidentaircraft was still on the fueling ramp and returned to Stead Airport and reported noproblem with the fuel.
A modest amount of burned luggage residue was found in the area of the nose baggagecompartment at the accident site. Two separate foot searches of the runway environs failedto locate any items from the aircraft. The husband of the pilot’s mother brought her tothe airport and helped load luggage. He reported through a friend of the family that thegroup only had about 20 pounds of soft luggage. He reported that he only helped load thebaggage and did not operate the baggage door.
No fuel sample was retained by the fuel pump operator (Texaco). However, refuelingrecords for the time following the refueling of the accident aircraft (attached), showthree other aircraft were fueled after the accident aircraft and no abnormalities werereported. The records show that the pilot added 119.6 gallons. The clock times on therecord are erroneous.
Personnel Information
The pilot was employed as a flight engineer at a supplemental air carrier. The pilot’slogbook was located in the wreckage with fire damage. It indicated total flying time of484 hours as of November 15, 1997. The pilot’s multiengine airplane rating was issued June26, 1997, after 7.2 hours of instruction and check ride in a Beech BE-95. On August 1,1997, the pilot logged 1.9 hours in a Piper PA-34-200 and 2.3 hours in a Beech BE-58. OnNovember 14 and 15, 1997, the pilot flew 8.6 hours during his checkout in the accidentaircraft.
One of the two flight instructors who performed the checkout in the accident aircraftsaid that they do not normally simulate engine failures on takeoff or do Vmcdemonstrations during checkout of a rated pilot. They did do slow flight, steep turns, andstalls, as well as a high altitude airport landing and takeoff. They also practiced engineout procedures by retarding the mixture control while in cruise flight. This instructorconsidered the pilot a "very good pilot" who flew well and kept the airplaneunder control.
There were no further entries in the pilot’s logbook after November 15. The SafetyBoard investigator, using a DUATS, no wind, flight plan, estimated that the pilot flew anadditional 3.1 hours in the accident aircraft while en route from Pomona, California, toSanta Rosa, California, to Carson City preceding the accident.
Aircraft Information
The aircraft’s weight and balance at the time of the accident was estimated by theSafety Board investigator using postaccident weights of the occupants provided by theCoroner’s office, together with baggage and fuel weights reported from other sources. Theinitial calculation indicated that the takeoff center of gravity position was 100.32inches aft of datum. However, a subsequent examination of the aircraft weighing recorddated October 29, 1997, revealed an error in the calculation which caused the empty weightcenter of gravity position to be reported 2.4 inches aft of the true position. Aftercompensating for this error, the calculation showed the takeoff weight of the accidentaircraft was 4,730 pounds and the center of gravity was at 98.5 inches aft of datum.According to the weight and balance documents, the maximum takeoff weight is 5,200 poundsand the center of gravity limits (at 4,730 pounds) are 95.3 to 100.5 inches.
In the operator’s "Flight Squawk Record" document, squawk number 2, datedSeptember 15, 1997, is "door ajar lt inop." The corrective action, datedSeptember 16, 1997, is "adjusted fwd baggage compartment door switch. Light worksgood." According to the operator, the door ajar light was located on the glareshieldin front of the right seat passenger.
Meteorological Conditions
According to a deputy from the Carson City Sheriff’s Department, weather at Carson Cityat the time of the accident was clear and sunny, calm wind, and temperature in the mid40-degree Fahrenheit range.
Wreckage And Impact Information
The aircraft impacted in the roof of a "Porta-Hangar" single aircraft,T-hangar located approximately 1,000 feet from the departure end of runway 27 and 200 feetsouth of the runway centerline. The hanger, in the center row of three rows of"Porta-Hangar’s" was identified as B6 and is located at latitude 39 degrees11.13 minutes north, and longitude 119 degrees 44.26 minutes west (GPS). There was a firefollowing impact with the ground and the entire aircraft was destroyed by the fire. Burneddebris associated with the aircraft was found on the floor of the hangar approximately inthe planform of the aircraft with the fuselage aligned approximately 300 degrees. A CessnaT210 aircraft was stored in the hangar and was also destroyed.
The fuselage was oriented with the nose in the northwest corner of the hangar and theskeletal remains of the empennage about 25 feet southeast near the center of the hangardoor. The aluminum components of the fuselage were consumed by fire except for the areaunder the cabin floor in proximity of the wing center section structure. The instrumentpanel, switches, and radios were destroyed by impact damage and fire. The cabin area wasidentifiable by the steel tube truss cabin structure, seat frames, and lower fuselage inproximity of the wing center section. The cabin floor beneath the pilot’s seat wasbuckled. The pilot’s seat remained attached to the right rail and was detached from theleft rail. The aft travel seat stop was in place on the left rail and the seat waspositioned forward of the stop. The engine power levers were bent to the right side. Thethrottle and propeller controls were approximately in the mid-range position with theleft-hand engine controls forward of the right hand engine.
The nose baggage door latch handle and recess assembly, and the latch operating shaftswere located in the nose area of the wreckage. The handle was approximately 1/4-inch openat the locking end and solidified molten metal had filled the cavity between the handleand the recess fixture. The track for the nose baggage door hold-open rod was foundcrushed with the slot in the track.
The landing gear were in the retracted position. The flight controls were continuousfrom the cockpit to the empennage and the aileron cables outboard of the nacelles had beendisturbed by rescue personnel. According to the representative from The New Piper AircraftCompany, the threads exposed on the trim tab actuators corresponded to neutral rudder trimand approximately 2 degrees nose down elevator trim. The flaps and flap actuator weredestroyed by impact and fire damage.
The wing center section, engine nacelle structure, and engines remained attached to thefuselage. The wings were consumed by fire outboard of the engine nacelles; however,residue on the hangar floor associated with wing approximated the wing planform. The fuelselector valves in the nacelles were positioned to the outboard tank position and wereheat damaged.
Both engines remained attached to their respective engine mounts and remained alignedwith the fuselage. The engine sumps and induction systems of both engines were destroyedby impact and fire. The right engine fuel servo assembly was destroyed by fire. Thethrottle valve in the left servo assembly, although damaged by impact and fire, was nearthe closed position. The fuel distribution manifold was intact on the top of the engines,however, no fuel was present. The engine fuel supply lines and engine mounted fuel pumpswere destroyed by fire. The exhaust systems were destroyed by impact. The magnetos andignition harness wires on both engines were damaged by fire.
Both propellers remained attached to their respective engines and were in the featheredposition. The actuator arms of both propeller governors, however, were in the low pitchposition. The blades of the left propeller were free to rotate in the hub and the bladesof the right propeller would not rotate in the hub.
The blades of the right propeller were bent aft approximately 15 degrees about 12inches inboard of the blade tip. The R1 blade of the right propeller exhibited one gougein the leading edge approximately 1/4 inch and no chordwise striations. Approximately 20percent of the R2 blade tip was consumed by fire. The leading edge inboard of the meltedportion contained three gouges of approximately 1/4 inch and light chordwise and diagonalstriations.
The L1 blade of the left propeller was bent aft about 90 degrees between the root andthe midspan location. The blade had two gouges in the leading edge, one approximately 1/4inch and the other of approximately 3/8 inch. The L2 blade was bent forward approximately60 degrees and 12 inches inboard of the tip. The blade had one gouge on the face side ofthe leading edge near the tip and approximately 1 to 2 inches of the blade tip trailingedge was absent.
According to a deputy of the Washoe County Sheriff’s Department who was present whenthe occupants were removed from the aircraft, the pilot’s mother and brother were seatedin the number 5 and 6 seats, respectively. The number 3 and 4 seats were empty and theother female passenger was in the right front seat. There was a modest amount of personalbelongings in the nose area and, in the cabin, only a pillow and a jacket were found.
The engines were disassembled and inspected by the Safety Board at the facilities ofTextron Lycoming on January 21 and 22, 1998. The sumps were consumed by fire on bothengines and the data plates were absent. The engines were identified by metal tags affixedat the accident site. The fuel, induction air, ignition, and exhaust systems could not beexamined due to impact and fire damage. The engines were mechanically continuous, thevalves functioned and accessory case gears rotated when the crankshaft was turned. The oilpump gears were intact. The internal mechanical components of the engines were in placeand undamaged except for the effects of heat generated by the postaccident fire. Thecrankcase and connecting rod bearings were gray and unscored, although tin plating on someof the bearings had melted and run in localized areas of high temperature. The connectingrod journals were smooth, shiny, and unscored, although several journals exhibited a blueappearance in areas of high temperature. The spark plugs were clean and gray except forthe number 1, 3, and 5 cylinder plugs on the left engine which were covered with shinyblack oil. The fuel injectors were clear except the number 1 injector on the left engineand the number 2 injector on the right engine which were blocked. The number 4 and 6injectors on the right engine were partially blocked. The oil filters were opened and theelements were charred. There were no metal particles in the filter elements. The propellergovernors, when removed from the engine, turned freely and pumped oil.
The propellers were disassembled and inspected by the Safety Board at HartzellPropeller Company on January 23, 1998. The dome and piston of the right propeller wereabsent. The pitch control rod of the right propeller was broken approximately midwaybetween the piston and the yolk. The fracture was elliptical and an area of displacedmetal on the forward face of the hub conformed to the fracture shape of the rod. Accordingto a Hartzell representative, the position of the rod failure relative to the forward faceof the hub corresponds approximately to a low pitch blade position at the time the markwas transferred. A mark transferred from the R2 blade root to the adjacent preload platecorresponded to a 21-degree blade position at the time the mark was transferred, accordingto the same Hartzell representative. The blade root pin of the R1 blade was bentapproximately 5 degrees toward the low pitch direction and the pin on the R2 blade wasbent a like amount in the high pitch direction.
The pitch control rod of the left propeller was broken in the threaded area on bothsides of the yolk assembly. The remaining portion of the rod aft of the yolk was bent atmid-length. According to the Hartzell representative, if the bent rod was positioned sothat the bent portion was between the yolk and the last point of support for the rod onthe aft face of the hub cavity, the blade position corresponded to approximately the lowpitch position when the rod was bent. Transfer marks on the preload plate of the L2 bladecorresponded to blade angles of 40, 78, and 88 degrees, according to the Hartzellrepresentative. The pitch change pins on the root of both blades of the this propellerwere bent approximately 5 degrees toward the high pitch position.
Additional Information
An autopsy was performed on the pilot by the Washoe County Medical Examiner/Coroner anda toxicology analysis was performed by the FAA Civil Aeromedical Institute.
The aircraft wreckage was released to Mr. Ernest Despain, insurance adjuster for Kernand Wooley, LLP, on February 11, 1998.
Probable Cause
The failure of the pilot to maintain airspeed after being distracted by the nosebaggage door opening in-flight resulting in a stall/spin. Factors contributing to theaccident were the pilot’s inadequate preflight inspection and lack of experience in thetype aircraft.
