FTW96FA151

The in-flight loss of a bolt in the elevator control system due to improper installation by unknown person(s), which resulted in a disconnection of the elevator control system that rendered the airplane uncontrollable.

Radar data indicated the airplane was maneuvering in a manner consistent with the performance of aerobatics prior to entering an uncontrolled descent during which it reached a speed of over 220 knots. Evidence indicated that, prior to impact, the pilot jettisoned the canopy, released his restraining harness, and deployed his emergency parachute. All of the suspension lines attaching the parachute canopy to the harness separated. The manufacturer recommended that the parachute not be deployed at speeds greater than 140 knots; however, this limit was not marked, nor was it required to be marked, on the parachute. Examination of the airplane's elevator control system revealed evidence that a bolt attaching a control tube to a bellcrank migrated out during the flight. The bolt should have been secured with a castellated nut and a cotter pin. The control tube end, bolt, castellated nut, and cotter pin were not recovered from the accident site. It could not be determined whether the bolt was improperly installed when the airplane was manufactured or had subsequently been removed and reinstalled incorrectly.

HISTORY OF FLIGHT On March 21, 1996, at 1741 central standard time, a Sukhoi SU-29, N729SU, was destroyed when it impacted terrain following a loss of control while maneuvering near New Orleans, Louisiana. The airline transport rated pilot, the sole occupant of the airplane, received fatal injuries. The airplane, co-owned and operated by the pilot, departed from Lakefront Airport in New Orleans at 1727 for the local personal flight conducted under Title 14 CFR Part 91. Visual meteorological conditions prevailed and a flight plan was not filed. During interviews, conducted by an NTSB investigator, the co-owner and a relative reported that, during the week of the accident, the pilot had made 3 or 4 flights per day to the local practice area (approximately 7 nautical miles southeast of the Lakefront Airport) to perform aerobatic maneuvers. The pilot's intentions were to become more familiar with how the Sukhoi handled and to prepare for an upcoming air show. For these flights, the pilot occupied the rear seat and enough fuel was placed in the main tank to permit about 35 minutes of flight time. Review of National Track Analysis Program (NTAP) and tracking radar data obtained from the Federal Aviation Administration (FAA) indicated that, after takeoff, the airplane proceeded southeast to the practice area. During the time period from 1732 to 1740, the airplane remained in the practice area and maneuvered between 3,900 and 1,800 feet MSL. At 1740:09, the airplane was at 3,400 feet MSL. The last NTAP radar return was recorded at 1740:19 and showed the airplane at an altitude of 900 feet MSL. Calculations performed using both the last two NTAP radar returns and the last two tracking radar returns indicated the airplane's average speed during the final seconds of the flight was over 220 knots (253 mph.) See the enclosed NTSB Recorded Radar Study for further details. There were no reported witnesses to the accident. Concerned relatives of the pilot reported the airplane missing and a search was initiated at 2120. The wreckage was located approximately 10 nautical miles southeast of the Lakefront Airport at 0815 on March 22, 1996. PERSONNEL INFORMATION The pilot and the co-owner purchased the airplane on February 1, 1996, from Pompano Air Center (PAC), Pompano Beach, Florida. On February 1st and 2nd, the pilot received 7.5 hours of ground training and 3.4 hours of flight training in N729SU from a PAC flight instructor and, subsequently, flew the airplane from Florida to Louisiana. According to the pilot's father, at the time of the accident, the pilot had accumulated approximately 18 hours flight time in the airplane. Prior to purchasing the Sukhoi, the pilot owned and flew other aerobatic airplanes including a Pitts S-1S. The pilot held a current Statement of Aerobatic Competency (FAA Form 8710-7) issued on December 18, 1995, for solo aerobatic flights with no altitude restriction in Pitts and Swift aircraft. AIRCRAFT INFORMATION The single-engine, tandem two-seat aerobatic airplane was manufactured on April 11, 1994, by Advanced Sukhoi Technologies (AST), Moscow, Russia. The airplane was assembled and test-flown in Russia, then partially disassembled (empennage, wing, and landing gear) for the purpose of shipping. The disassembled airplane was shipped to PAC, which is the worldwide distributor of Sukhoi aerobatic airplanes. At PAC, the airplane was reassembled and application was made to the FAA for a special airworthiness certificate in the experimental category for the purpose of exhibition. An entry in the airframe logbook dated October 17, 1995, stated the following: AIRCRAFT RE-ASSEMBLED AFTER SHIPMENT FROM RUSSIA. FLIGHT CONTROL RIGGING CHECKED. BRAKES BLED. FUEL SYSTEMS LEAK CHECKED. ALL PROCEDURES CONDUCTED IN ACCORDANCE WITH SUKHOI SU-29 MAINTENANCE MANUAL. "I CERTIFY THAT THIS AIRCRAFT HAS BEEN INSPECTED AND IS IN A CONDITION FOR SAFE OPERATION." The experimental airworthiness certificate was issued on October 19, 1995. An entry in the airframe logbook dated February 1, 1996, indicated that a Russian communications radio was removed and a transponder, encoder, navigation/communication transceiver, and navigation indicator were installed. There were no further entries in the logbook. During an interview, conducted by the investigator-in-charge, the co-owner was questioned about maintenance performed on the airplane after its purchase from PAC. He reported that the only maintenance he was aware of involved an incident in which the pilot had experienced an in-flight restriction of control stick travel. After making an uneventful landing, the pilot, who was an FAA-certificated mechanic, examined the airplane's control system and found that a radio had slid from its case and was interfering with the controls. The pilot secured the radio with safety wire and reported no further problems with the control system. According to the co-owner, at the time of the accident, the airplane had accumulated approximately 20 hours flight time. The elevator control system of an SU-29 consists of tubes, a bellcrank, and intermediate joints that connect the control sticks to the elevator for pitch control (see enclosed diagram). The bellcrank, which mounts directly behind the rear seat, is oriented vertically, and tubes are attached to the bellcrank in the fore and aft directions. Bolts, secured with castellated nuts and cotter pins, are used to attach the tubes to the bellcrank's clevis tangs. The lower tube runs forward and connects the bellcrank to the aft cockpit control stick. The upper tube runs aft and connects at a hanging bracket to another control tube which connects to the elevator. WRECKAGE AND IMPACT INFORMATION The accident site was in a sparsely wooded swamp approximately 29 degrees 57.8 minutes north latitude and 89 degrees 53.1 minutes west longitude at sea level elevation. All wreckage was located within approximately 90 feet of the central impact crater. The water-filled crater measured about 10 feet in diameter and was flanked on the west by fragmented pieces of the left wing and on the east by fragmented pieces of the right wing. The elevator tips were embedded in the mud on the north edge of the crater with the left elevator tip located west of the right tip. During excavation of the crater, the pilot's body and fragmented pieces of the airplane were recovered from depths of 15 to 24 feet. Recovered parts included major portions of the wing, empennage and fuselage, both main landing gear, propeller blade fragments, and pieces of the engine exhaust system. Among the parts not recovered were the engine, the front and rear cockpit instrument panels, and the radios. On March 28 and 29, 1996, a wreckage reconstruction, supervised by the NTSB investigator-in-charge, was performed. Examination of the recovered portions of the wings and empennage did not disclose any evidence of preimpact structural failure or separation. A layout of the recovered portions of the fuselage revealed that the front cockpit was destroyed and the rear cockpit sustained substantial deformation. The front seat was destroyed; however, the rear seat remained intact and attached to its anchor points. Damage to the rear seat was confined to the upper left quadrant of the seat back which was penetrated and deformed. The front 5-point restraining harness was fastened and the rear harness was unbuckled. The airplane's canopy, which is designed to be jettisoned by the pilot during an emergency bail out, was not recovered from the accident site. The elevator control system tubes were broken in numerous places. The bolted connections in the elevator control system remained secure with the exception that the upper (aft) tube was detached from the elevator bellcrank. There was no visible bending of the upper clevis tangs of the bellcrank and no visible deformation of the upper clevis cross bolt holes. The forward portion of the upper tube, as well as the bolt, castellated nut, and cotter pin used to attach it to the bellcrank, were missing. The aft portion of the upper tube remained bolted to the hanging bracket. MEDICAL AND PATHOLOGICAL INFORMATION An autopsy of the pilot was performed on March 25, 1996, by Dr. Paul McGarry at the Orleans Parish Coroner's Office. Toxicological tests were negative for drugs. According to Dr. Canfield of the FAA's Civil Aeromedical Institute, the toxicological finding of ethanol (alcohol) in heart and vitreous fluid was probably from postmortem ethanol production and "would not typically be considered a factor in this accident." SURVIVAL ASPECTS The canopy of the pilot's emergency parachute was found 35 feet from the central impact crater on a magnetic heading of 120 degrees. All of the 20 suspension lines joining the canopy to the harness had separated. (Sixteen of the lines separated at the canopy's lower lateral band, three lines were torn from the canopy in their entirety, and one line separated 2.5 inches above the harness connection point.) The pilot chute, a small parachute which is attached to the apex of the canopy and deploys first when the ripcord is pulled, was torn apart. The base of the pilot chute remained attached to the canopy, and the separated section was found 20 feet from the central crater on a magnetic heading of 050 degrees. The harness, container and ripcord of the pilot's emergency parachute were provided to the NTSB by the coroner. According to the autopsy report, the separated suspension lines were "tangled around the body," the harness was "intact and in place," and the "metallic rip cord [was] pulled loose at the end." At the request of the NTSB investigator-in-charge, on April 12, 1996, all components of the emergency parachute were examined by 2 FAA inspectors and 1 FAA engineer, all certificated parachute riggers. See the enclosed report for their detailed description of the damage to the parachute. The parachute was a National 360, Model NP6-1, manufactured by National Parachute Industries of Flemington, New Jersey, in April 1985. The packing card indicated the last repack was performed on August 5, 1995, at the manufacturer's facility in Flemington. Both the parachute canopy, which was identified as a Phantom 24, and the container were marked with the statement: "MANUFACTURED IN ACCORDANCE WITH FAA TSO-C23b." Technical Standard Order C23b (TSO-C23b) specified that "parachutes manufactured for use in civil aircraft of the United States on or after March 29, 1962, shall meet the minimum performance standards of National Aircraft Standards Specification 804 [NAS 804]." Records provided by the FAA indicated that National applied for TSO authorization for the NP6-1 parachute assembly on June 5, 1981. According to the President of National, this was the first time the company had applied for a TSO authorization. Included with National's letter of application was a one page document entitled "Statement of Conformance," which read in part: "This is to certify that the NP6-1 and NP6-2 parachute assemblies meet the requirements of TSO-C23b and NAS 804 as referenced in FAR parts 21 and 37." On July 1, 1981, the FAA authorized National to mark the NP6-1 parachute assembly "with the applicable TSO-C23b marking." The FAA's letter of authorization stated in part that "the data and certification of conformance . . . forwarded with your letter dated June 5, 1981, have been found acceptable." FAA Order 8150.1, issued May 19, 1970, and valid until superseded on September 21, 1987, contained instructions for use by FAA inspectors in administering the TSO compliance program. Paragraph 4.d provided the following guidance for evaluation of applications for TSO authorization: It is recommended, particularly in the case of new TSO manufacturers, that an engineer visit the applicant's facility for the purpose of appraising the applicant's competence to certify conformance with the TSO. The engineer should assure himself that compliance tests, as prescribed, are being realistically conducted. This visit should take place, preferably within the 30-day period before an authorization is granted. No documentation of any visits made to National prior to granting the TSO authorization was contained in the FAA's response to an NTSB request, dated July 23, 1996, for "copies of all records, files, correspondence, memoranda, manuals, etc., in the possession of the FAA concerning TSO authorizations granted to National." One of the compliance tests required by NAS 804, Paragraph 4.3.6.1, for a standard type parachute was a strength test consisting of three drops "made with a dummy weight and indicated airspeed to give the equivalent of 5,000 lbs. shock load." NAS 804 contained a table listing 10 combinations of launching speeds and weights calculated to produce a 5,000 pound shock load on a 28 foot standard flat-type parachute (commonly known as a C-9 parachute.) The table indicated that, when using a launch speed of 150 mph, a dummy weight of 660 pounds was required to develop a 5,000 pound shock load. For a launch speed of 250 mph, the required dummy weight listed was 275 pounds. Correspondence received from National indicated that the figures contained in the NAS 804 table were not applicable to the Phantom 24 parachute canopy, because the size, shape, and fabric of the Phantom 24 were different from those of the C-9. National calculated that a dummy weight of 362 pounds dropped at an indicated airspeed of 160 mph was required to develop a 5,000 pound shock load on a Phantom 24 canopy. Review of these calculations revealed that they relied on an assumed value for canopy inflation (fill) time. The formula National used to compute fill time was "a conservative assumption" drawn from "notes from [a] 1982 lecture" by the author of several parachute design manuals. This author had "not done actual tests" on the type of fabric used in the Phantom 24 canopy. National performed the strength drop tests on April 25, 1981, using their calculated weight of 362 pounds and launch speed of 160 mph. No measurements of the shock load developed were made. According to National, "the figures used to determine the 5,000 pound shock load were discussed with various FAA personnel both during and after the TSO testing process." No record of these discussions was contained in the FAA's response to the NTSB request of July 23, 1996. National pointed out that NAS 804 contained "no operational limits for parachutes tested in accordance with 4.3.6.1 (a parachute can be used at weights and airspeed above those at which it was tested)." National's emergency parachute manual, issued in 1984 and in effect at the time the accident parachute was manufactured, did not contain an operating limitations section. The revised manual issued on January 1, 1991, contained the following operating limitations applicable to the National 360: Recommended minimum deployment height: 500 ft AGL Recommended pilot weight: 100 - 177 lb Normal altitude loss during opening: 200 - 300 ft Opening time: 2 1/2 - 3 seconds (normal opening) Opening speed: 140 knots maximum During his most recent FAA flight physical on April 19, 1995, the pilot's weight was recorded as 162 pounds. NAS 804 required that the parachute pack and canopy be permanently marked with the following information: manufacturer's name, model number or model name, parachute serial number, date of manufacture, and National Aircraft Standard Number (NAS 804). According to National, for parachutes manufactured prior to February 1987, in addition to the required markings, a warning label was stamped onto each Phantom 24 canopy which stated the following: MAXIMUM GROSS WEIGHT* 188 lbs (85 kg.) MAXIMUM RECOMMENDED DEPLOYMENT VELOCITY 160 mph (139 kts.) at 188 lbs gross 170 mph (148 kts.) at 164 lbs gross 180 mph (156 kts.) at 141 lbs gross *GROSS WEIGHT INCLUDES JUMPER AND ALL EQUIPMENT During examination of the accident parachute, the required markings were found; however, no warning label was found. For parachutes manufactured during or after February 1987, National began sewing a warning label to each Phantom 24 canopy which read, in part: LOWER THE RISK OF DEATH, SERIOUS INJURY, CANOPY DAMAGE AND HARD OPENINGS BY NEVER EXCEEDING THE LIMITS SHOWN BELOW: MAXIMUM DEPLOYMENT SPEED 140 KNOTS MAXIMUM GROSS WEIGHT (JUMPER + 188 POUNDS CLOTHING + EQUIPMENT) TESTS AND RESEARCH The elevator bellcrank assembly, the recovered piece of the upper control tube, and the rear seat were sent to the NTSB Materials Laboratory in Washington, D.C., for examination. Measurements of the upper bellcrank clevis cross bolt holes using an optical comparator and a video measurement system confirmed that the holes were round and undistorted. Examination of the holes under higher magnification revealed that a bolt, washer, and nut had been previously installed with the bolt inserted from right to left (nut on the left side). The damage to the rear seat frame was found to be "consistent with contact by an object coming from behind the seat." Bluish green paint transfers matching the paint color of the control tubes were noted on the medium gray seat pan and frame. For further details see the enclosed NTSB Metallurgist's Factual Report. Sections of the wing spar were also sent to the NTSB Materials Laboratory. Examination of the upper and lower wing spar caps revealed a solid composite structure with no evidence of defects. ADDITIONAL INFORMATION On April 22, 1996, NTSB investigators examined three partially assembled SU-29 airplanes at PAC and found inconsistency in the assembly of bolts, castellated nuts, cotter pins, and elevator control tubes on these airplanes. On one airplane, the cotter pin on the upper bolt, which attaches the forward control stick to the aileron control tube, was installed but not flared. Some of the other cotter pins in the fuselage frame were partially flared. During a discussion with NTSB investigators, PAC maintenance personnel stated that a detailed inspection was not performed before a Sukhoi airplane was reassembled at PAC. PAC's normal procedure was to inspect only the work that it performed during reassembly of the wing, empennage, and landing gear. An inspection of the work performed at AST in Russia was not performed. PAC personnel further stated that no checklists, process sheets, or other guidelines were available to help them in the assembly and inspection of the airplanes. PAC mechanics were factory trained in the assembly of Sukhoi aerobatic airplanes and used the manufacturer's maintenance manuals to guide them in reassembling the airplanes. In a letter to NTSB staff dated June 3, 1996, the President of PAC stated that "PAC mechanics complete a detailed inspection of the entire aircraft prior to local test flight, FAA certification and delivery to the customer." He further stated that "the reassembled portions of the [accident] airplane were intact and inspected prior to leaving our facility, including the control rod, bolts, nuts and cotter pins." In a letter to the NTSB investigator-in-charge dated November 22, 1996, the President of PAC reported that during the pilot's flight training at PAC, the pilot "had difficulty reaching full forward control stick travel during some maneuvers." The PAC flight instructor informed the pilot that he had experienced "a similar problem in his own aircraft, and had the elevator control linkage adjusted to reposition the control stick so that a full range of travel could be reached." The pilot did not choose to have this adjustment made at PAC, but did state that as an FAA-certificated mechanic, he "could make the adjustment on his own." According to the President of PAC, "the bolt which was allegedly missing prior to impact is the very one which would have been removed to make such an adjustment." When questioned by the investigator-in-charge, the co-owner of the airplane and the pilot's father stated that they had no knowledge of the pilot making any adjustment to the elevator control linkage. FAA Order 8130.27, issued January 11, 1996, provides guidance to FAA inspectors in the issuance of special airworthiness certificates for the experimental purpose of exhibition. Appendix 1, paragraph 2.b.(7), of this order requires the inspector to verify that the applicant for a special airworthiness certificate "has made an entry in the aircraft records to certify that an appropriately rated FAA-certificated mechanic has accomplished the applicable inspection (as defined in paragraph 8 of this order) within 30 days prior to the submittal" of the application. Paragraph 8.a.(4) of the order states that performance competition aircraft (such as the SU-29) "must be inspected each year in accordance with an inspection plan that contains the scope and detail of 14 CFR Part 43, Appendix D." On April 23, 1996, NTSB investigators interviewed the FAA airworthiness inspector who issued the special airworthiness certificates for N729SU and other Sukhoi airplanes reassembled at PAC. He stated that he conducted a limited inspection of each airplane, including verification of control continuity from the control surfaces to the control sticks, before issuing the airworthiness certificate. He also stated that he examined each airplane's logbook for a signed statement by an FAA-certificated mechanic certifying the airplane's condition for safe operation. The inspector's interpretation of FAA Order 8130.27 was that a condition inspection in accordance with the scope and detail of Appendix D of 14 CFR Part 43 was not required for a newly manufactured airplane prior to the issuance of a special airworthiness certificate. On April 5, 1996, a letter was issued by AST to all SU-29 and SU-31 owners, recommending "inspection of ailerons, elevator, rudder, and rudder trimmer controls" to ensure "connections reliability and presence of locking devices." On April 23, 1996, a letter was issued by PAC to all SU-26, SU-29, and SU-31 owners providing preliminary accident findings and forwarding the AST letter to advise Sukhoi owners to recheck proper cotter pin installation and safety wiring of bolts and nuts. (The SU-26 and SU-31 are single-seat aerobatic airplanes manufactured by AST.) On May 31, 1996, the NTSB issued the following safety recommendations to the FAA: A-96-23 Issue a telegraphic airworthiness directive to the owners of all U.S.-registered Sukhoi SU-29 and SU-31 airplanes requiring immediate inspection to ensure that the flight control systems are properly assembled and that all control system cotter pins are properly installed. A-96-24 Amend Federal Aviation Administration Order 8130.27, Appendix 1, paragraph b.(7), to specify that applicable airplanes must be inspected in accordance with an inspection plan that contains the scope and detail of 14 CFR Part 43, Appendix D, before issuance of special airworthiness certificates. Also, ensure that the aircraft record entry made by the FAA-certificated mechanic, before issuance of a special airworthiness certificate, indicates that the aircraft meets the inspection standards of 14 CFR Part 43, Appendix D. On August 7, 1996, the FAA responded to these recommendations and indicated the following: A-96-23 Airworthiness directives (AD) are issued when an unsafe condition is established related to type design. Since the FAA has not issued a type certificate for the Sukhoi SU-29 and SU-31 airplanes, AD action cannot be supported against these airplanes. As an alternate action, on June 11, 1996, the FAA issued a Special Airworthiness Information Bulletin (ACE-96-03) to all U.S.-registered owners advising them of the need to conduct a complete inspection of the control systems as recommended by the Board. A-96-24 FAA will issue, by October 1996, Guidance Memorandum 96-12 to manufacturing inspection and flight standards district offices to provide guidance concerning the procedural aspects and record entry requirements to be used to issue special airworthinesss certificates. The information contained in the guidance memorandum will be included in the next revision to Order 8130.27. The retained aircraft records were returned to the co-owner on April 25, 1996, and the recovered wreckage and the parachute were released to the co-owner and the estate of the pilot on July 1, 1996. -