NTSB Factual Report

HISTORY OF FLIGHT

On May 28, 2000, at 1159 hours Pacific daylight time, a Piper PA-46-310P, N567YV, collided with the ground in a shopping mall parking lot while maneuvering to return to the runway during the takeoff initial climb from the Hawthorne, California, airport. The airplane, owned and operated by the pilot under 14 CFR Part 91, was destroyed in the collision sequence and post crash fire. The private pilot and two passengers sustained fatal injuries. Visual meteorological conditions prevailed for the personal flight and no flight plan was filed. The flight was originating at the time of the occurrence as a cross-country flight to Las Vegas, Nevada.

The accident sequence was heard and observed by witnesses both on the airport and in the vicinity of the crash site.

The airport witnesses, who knew the pilot and the airplane, hold various pilot certificates, and were on the ramp at different locations around the airport. All described the engine sounds during the takeoff as abnormal in various degrees. One witness said that while the sound was "different" than he had heard coming from this airplane before, it was not unusual enough to make him stop polishing his airplane. When the witness did look at the departure end of the runway, he observed that the airplane's climb out angle was "much shallower" than he had ever seen this airplane climb before, and he believed the takeoff ground roll had been longer than usual. Two other witnesses said the engine sounded "like a radial engine" and both believed that the power output was lower than normal. These two witnesses also observed a shallow climb out angle as the airplane flew over the west boundary fence.

Additional airport witnesses included the aircraft mechanics at the Northrup-Grumman executive hangar, which is located near the mid point of the runway on the south side and almost opposite the Federal Contract Control Tower. The maintenance department's chief inspector stated that he saw the airplane on the takeoff roll, and heard the engine surging. He stated that the propeller tone did not change from a steady rpm, but that the engine was "not developing good power output" and just did not sound like it was at full power. He said, that in his experience, he associated it with a fuel feed problem, a possible turbocharger surge, or an excessively lean running condition. The engine was not missing. The exhaust tone would alternatively get loud, then fade out, and become loud again. He said that the airplane was about midfield when it rotated (about 3,000 feet down the runway). It appeared to him that the airplane had not accelerated to takeoff speed, but the pilot forced the rotation by "yanking it off the ground." He also observed that the takeoff initial climb appeared mushy. He watched the climb out to about 20 feet and then went back to work.

The ground witnesses located near the impact site were uniform in their description of the airplane beginning a steep left turn between 1/4- and 1/2-mile from the runway's end. Those witnesses, who are residents of the area, stated that the airplane's altitude was much lower during this turn than normally observed with other airplanes that are departing from the airport. The bank angle was variously estimated by the different witnesses as 45 degrees or greater. The turn did not stop after 90 degrees like most other airplanes departing the airport, but continued around until the nose suddenly dropped and it entered a spiraling descent to ground impact in the shopping mall parking lot. The majority of these witnesses stated that they heard "sputtering" or "popping" noises coming from the airplane.

The self-service automated fuel pump at the airport recorded that the pilot swiped his credit card through the machine reader at 1135:31 to turn the pump on (See AIRCRAFT INFORMATION section of this report for complete details on the fueling of the aircraft). According to the air-ground communications recording at the Hawthorne Federal Aviation Administration (FAA) Contract Tower (FCT), the pilot contacted ground control at 1153:12, stated that he had the ATIS information, and requested a clearance to taxi for takeoff. The ground controller cleared the aircraft to taxi and hold short of taxiway delta for crossing instructions. The controller subsequently cleared the aircraft to cross the runway and taxi to the approach end and the pilot acknowledged this clearance at 1153:58. The pilot reported to the local controller that he was ready for takeoff at 1157:33, and requested a downwind departure. He was issued a takeoff clearance on runway 25 with a left downwind departure at 1157:39, and the pilot acknowledged the clearance at 1157:46. Between 1157:49 and 1157:57, the pilot and the local controller exchanged transmissions about the pilot's trip to Las Vegas. By referencing a scale airport diagram, the approximate taxi distance from where the aircraft crossed the runway to the approach end of the runway is 2,400 feet. The run-up area for runway 25 is near the departure end of the runway. No distress calls were recorded on the local or ground control frequencies.

Recorded radar data in the form of primary targets and 1200 code secondary beacon returns were obtained from both the Los Angeles Air Route Traffic Control Center long range search radar antenna near San Pedro, and from the Southern California TRACON ASR-9 south-side antenna site at the Los Angeles International Airport. The primary and secondary beacon returns for this aircraft were isolated on the basis of the known takeoff time and the fact that no other aircraft were cleared for takeoff within 5 minutes either side of the accident aircraft. The targets were then plotted on a USGS 7.5 minute chart. The TRACON ARTS radar system computes average ground speed for secondary beacon targets. The wind direction and speed was recorded in the tower at the time of the accident as from 250 degrees at 9 knots. The first secondary beacon return from the west bound departing aircraft was recorded abeam the control tower with a mode C altitude report of 100 feet (the field elevation is 66 feet msl) and did not have a ground speed. The second beacon return was plotted 4 seconds later still over the runway about 700 feet east of the west boundary fence with a mode C altitude report of 200 feet and a computed ground speed of 74.7 knots. The third beacon return 4 seconds later was located about 750 feet west of the airfield boundary fence with a mode C altitude report of 200 feet and a computed ground speed of 74.7 knots. The fourth beacon return was about 1,500 feet west of the third return, with a mode C altitude report of 300 feet and a computed ground speed of 75.3 knots. The last recorded return was about 2,100 feet south-southwest of the fourth beacon return with a mode C altitude report of 400 feet and a computed ground speed of 82.0 knots. The last beacon return was noted to be about 800 feet west of the impact location.

The angle of bank versus stall speed chart for the aircraft in the Pilot's Operating Handbook was reviewed. For an airplane in a gross weight and a flaps and gear up condition, the stall speed in a 45-degree bank is 82 knots. For 60 degrees of bank, the stall speed rises linearly to 96 knots. The chart is appended to this report.

Engineering personnel at Piper Aircraft were asked by Safety Board investigators to compute a performance capability profile for a normal takeoff and climb. In the interests of conservatism, the analysis assumed a gross weight condition and a standard sea level atmosphere. The engineering analysis predicted that a normal ground roll would consist of 1,760 feet, with liftoff occurring at 77 knots indicated air speed (KIAS). The aircraft should have achieved 89 KIAS crossing runway end with an initial gear-down and flaps-up climb speed of 808 feet per minute. Normal climb speed is 110 KIAS, which should have been achieved at 200 feet agl, and the normal flaps and gear-up rate of climb would be 1,140 feet per minute.

PERSONNEL INFORMATION

Certified copies of the pilot's FAA airman and medical certification files were obtained for review. According to the files, the pilot held a private pilot certificate with airplane ratings for single engine land and instruments. The most recent issuance of the certificate was dated May 25, 1993, when the pilot passed his instrument airplane practical flight test. A third-class medical certificate was issued on May 13, 1999, with the limitation that he must possess correcting lenses for near vision while exercising the privileges of his airman certificate.

No personal flight records were recovered for the pilot. On the application for his May 13, 1999, medical certificate, the pilot stated that his total flight experience was 1,900 hours. On a May 20, 2000, application for the renewal of his aviation insurance policy, the pilot stated that he had 2,550 total hours, with 1,250 in the PA-46-310P. According to FAA aircraft registry records, the pilot has owned the airplane since October 1990.

At the request of Safety Board investigators, The New Piper Aircraft, Inc., reviewed factory records of the formal pilot training courses offered for the PA-46 series airplanes. No records were found that the pilot attended any formal factory course from 1997 through the date of the accident. According to Piper, a training vendor change occurred in 1997, and the previous company is no longer in business and no records exist prior to 1997.

Information provided to Safety Board investigators led to the identification of a flight instructor who endorsed a biennial flight review for the pilot in May 1999. The instructor stated that he had been an instructor with the Piper factory PA-46 training program throughout the 1980's. The instructor reported that he had provided recurrency flight and ground training to the pilot for the last 5 years. The most recent session was in May 1999, and included 7 hours of ground instruction and 4 hours of flight instruction in the pilot's airplane. At the conclusion of the sessions, the instructor endorsed the pilot's logbook for an instrument competency check, a biennial flight review, and the recurrency training required by the pilot's insurance policy.

AIRCRAFT INFORMATION

The aircraft's maintenance records were presented by family members to Safety Board investigators for review following a search of his business and residence for the records. In addition, the Aircraft and Aircraft Registry files maintained by the Federal Aviation Administration at the Aircraft Registry in Oklahoma City, Oklahoma, were reviewed for historical information.

Review of these records disclosed that the airframe, serial number 46-8408016, was manufactured by the Piper Aircraft Corporation in January 1984. The pilot purchased the airplane on September 19, 1990. The two most recent annual inspections were accomplished on August 21, 1999, and July 1, 1998, at total airframe times of 2,946 and 2,746 hours, respectively. The records note that Airworthiness Directives applicable to the airframe were complied with through 99-15-04. The recording hour meters in the airplane were destroyed in the accident sequence and no information was available on the total time as of the accident.

The Teledyne Continental TSIO-520-BE1B engine, serial number 273651-R, was remanufactured by the factory on October 29, 1990, and was installed in the airframe on December 21, 1990, at an airframe total time of 1,813 hours. The two most recent annual inspections corresponded to those recorded for the airframe; the engine total time recorded on the August 20, 1999, annual inspection was 1,132 hours. During this last annual inspection, the No. 3 cylinder was removed for overhaul, and subsequently reinstalled on the engine. New magnetos and a new ignition harness were also installed at that time. On September 1, 1997, all six cylinders were removed and then replaced with overhauled units.

The records also disclosed that the airplane had been modified with the installation of several items of equipment under Supplemental Type Certificate (STC) approval authority. The following is a list of the major modifications, the date installed, and the STC number:

1. Graphic Engine Monitoring System; 12/21/90; STC SA157NE 2. Wing SP9000 Spoilers; 03/18/92; STC SA5196NM 3. Increased Fuel Capacity Filler Ports; 12/19/95; STC SA00859AT 4. MT MTV-14-D 4-bladed wooden propeller and spinner assembly; 11/21/98; STC SA01542AT.

Concerning the propeller modification, the STC holder supplied the Safety Board investigators with the following information. The European certificated constant speed propeller is manufactured in Austria by MT-Propeller Entwicklung GmbH & Company and has been issued a US Type Certificate (P3BO). It consists of a conventional metal hub and hydraulically actuated pitch change mechanism coupled with four wooden core blades. The blades have integral steel rods incorporated into the metal blade butts that are in turn secured by conventional clamps into the hub. The internal steel rods in the blades are surrounded by glued-up wooden bodies that provide the blade shape and contour. The wooden blade bodies are further covered by thin composite skin and have a stainless steel leading edge abrasion strip. The diameter is about 70 percent of the 2-bladed metal propeller originally installed, and the principal advertised application of this propeller is a reduction in noise signature coupled with an increase in cruise performance.

Piper Aircraft Corporation was asked if they had any information on the MTV-14 propeller installation on the Malibu/Mirage series. According to Piper, in 1998 they evaluated the propeller for installation in the PA-46-350 Mirage and performed some limited cruise and climb performance tests. The MTV-14 did not perform as well as the standard propeller in some conditions. Takeoff performance tests were not performed. Piper said they elected not to use the MTV-14 propeller due to concerns over oxidation/corrosion of the leading edge strip, and the lack of firm data on the lightening and icing protection capability of the propeller.

According to Piper Aircraft, the fuel system for the PA-46-310P consists of wet-wing fuel tanks, one in each wing, with a capacity of 61 US gallons each, for a total of 122 gallons. One gallon in each tank is unusable. The aircraft maintenance records, the FAA Aircraft Registry file, and physical examination of the wreckage disclosed that it had been modified on December 19, 1995, with the installation of fuel filler ports outboard of the factory ones on each wing under STC SA 00859AT; this modification added 10 US gallons to the capacity of each tank.

Fueling facility records at the Hawthorne airport revealed that the airplane was fueled prior to takeoff with the addition of 120.2 gallons of 100LL aviation gasoline. According to the fueling facility at Hawthorne, the pump used by the pilot is a self-service pump, where you swipe a credit card to enable the pump. The time stamp on the receipt, 1135:31, is the time he swiped his credit card through the reader to turn on the pump. The fueling facility reported that the pump utilized by the pilot to refuel the airplane has a maximum discharge capacity at the nozzle of 24 gallons per minute. Other airplanes fueled before and after the accident airplane completed flights without difficulty.

According to family members and employees of the pilot's business, he had taken the airplane in December 1999 on a trip to points in the Caribbean and had returned after Christmas. No one could recall any flight activity by the pilot in the accident airplane between December and the accident flight. Security Aviation at the Hawthorne airport is the only fuel concessionaire on the field. A search of their records disclosed that the first record of the accident aircraft refueling prior to the accident flight was on December 21, 1999, when the pilot purchased 87.4 gallons. Safety Board investigators telephonically contacted all known aircraft fueling facilities in the greater Los Angeles area requesting information on any fuel purchases between December

NTSB Probable Cause

A partial loss of power due to water contamination in the fuel system and the pilot's inadequate preflight inspection, which failed to detect the water. The pilot's failure to perform an engine run-up before takeoff is also causal. Additional causes are the pilot's failure to maintain an adequate airspeed margin for the bank angle he initiated during the attempted return to runway maneuver and the resultant encounter with a stall/spin. Factors in the accident include the pilot's failure to detect the power deficiency early in the takeoff roll due to his diverted attention by a nonpertinent personal conversation with the local controller, and, the lack of suitable forced landing sites in the takeoff flight path.