FlightGlobal.com
Home
Premium
Archive
Video
Images
Forum
Atlas
Blogs
Jobs
Shop
RSS
Email Newsletters
You are in:
Home
Aviation History
1961
1961 - 0225.PDF
FLIGHT, 17 February 1961 225 LESSONS OF A TURBOPROP INQUEST BJ LOW are abstracts from the FAA report on the most intensiveaccident investigation in US aviation history, after major accidents to Lockheed Electras at Buffalo, Texas, in September 1959, and atOinelton, Indiana, in March 1960. The second crash was the outcome of << structural failure almost identical to that suffered by the first aircraft. FOLLOWING a meeting with the president of each airlineoperating Electras, Lockheed and Allison, the FederalAviation Agency initiated a five-point programme: —1) Emergency airworthiness, regulations. Reduction of the originally certificated normal operating speed of 373 miles perhour to 259 miles per hour. This speed restriction provided an additional airload margin of 50 per cent between the maximumloads expected in normal operation and the structural capability of rhe airplane, and a more rapid slow-down to rough-air penetra-tion speeds. The autopilot was deactivated until it could be proven that it was blameless. Pressure-refueling procedures were stressedto prevent straining by overpressurization of the tanks. (2) Emergency maintenance inspection. The FAA inspectedall Electra wings and daily checks were made of engine reduction gears. After severe turbulence or hard landings a thorough struc-tural inspection was given before flight was again permitted. (3) Special air carrier operations inspections. Emphasis wasplaced on instruction of flight crews in the new speed limitations. (4) Flight recorders. The FAA required that flight recorders—giving a permanent record of airspeed, altitude, direction, and acceleration—were installed in all Electras. (5) Engineering and research, investigations. Lockheed wasdirected by the FAA to undertake a program re-evaluating the Electra design. This included a flight-test program to demon-strate the soundness of the airplane with respect to manoeuver and gust loads. Particular emphasis was given to the ability ofthe wing to dampen after it was disturbed and loaded by air gusts. The control system was examined to determine the effectof malfunctions, and basic engineering data on strength and stress analysis was reviewed. An accelerated flight test program of a highly instrumentedElectra was started, to obtain data to demonstrate aerodynamic and structural characteristics of wings, nacelles, and control surfaceassemblies. Manoeuvers, gust penetration, and flutter response tests were conducted over a range of weights, speeds, and fuel loadssufficient to ensure that any hitherto unsuspected abnormal struc- tural effects present would be revealed. The following conclusions were drawn : (1) The original designmanoeuver loads were correct. (2) During an abrupt pull-up stability characteristics are normal. (3) No abnormalities arepresent in manoeuvers such as rolling pull-ups. The Electra was flown through heavy turbulence to measurewhat affect gusts had on the wing and fuselage structure. This gust flight program was most extensive and far-reaching. Newknowledge was gained about analyzing airplane stresses. Lockheed was directed to evaluate the control system and auto-pilot in respect of possible malfunctions, failures, and induced effects to insure that sudden destructive forces were improbable.As a result, FAA approved a series of changes to the control system and autopilot. Proving the Structure The most important part of the analysis was that related to proofof the structure itself. The extensive flight test program made it possible to validate estimated aerodynamic loads with actual loads.This process permitted a reaudit of the structure, including: 1) Integration of flight program data. (2) Review of existingstatistical data on magnitude, frequency, and characteristics of Electra gusts. (3) Re-investigation of flutter and divergencyanalysis, including consideration of abnormal situations arising as a result of failures such as damaged gear box support struc-tures, complete loss of one engine, and damaged wing structures. (4) Wind tunnel investigations to obtain aerodynamic data bothfor normal configurations (including propeller-induced effects on airflows, forces, and other phenomena, and including varied pro-peller inflow angles) and for abnormal conditions arising as a result of structural damage or failure. As a result of these studies andtests it was concluded that under certain conditions of engine nacelle or powerplant damage, a phenomenon, known as "whirlmode," could occur. As a wing is loaded intermittently (as might occur when encoun-tering gusts) it must return automatically to its previous position. Good damping requires that a motion of this sort die out rapidly.The damping forces are those which take energy away from the oscillations. A small amount of damping comes from internalenergy absorption in the structure and energy absorbing parts such as rubber engine mounts, but the greatest damping comes fromdeigning the structure so that the aerodynamic forces on the airplane resist the motions and thus absorb energy from the oscil-lations. Conversely, if the air forces aid the motion, the oscillation grows—this condition is known as flutter. The Electra wing has good positive damping, but the studiesconducted during the re-investigation show that while the control system could not induce wing oscillation, the wobbling of anoutboard propeller could. If there are certain structural failures in the outboard powerplant installation, the stage could be set fora wobbling propeller condition. Since a propeller has gyroscopic characteristics, it will tend tostay in its plane of rotation even though it may be loosely mounted. It will hold its plane of rotation until displaced by some strongexternal force such as atmospheric turbulence, abrupt aircraft manoeuvers and sudden power surges. When such a force isapplied, the propeller reacts 90° out of phase with the applied force. Assuming the propeller is displaced upwards, the struc-tural resistance of the mounting system applies a nose-down pitching moment. The propeller disc, when viewed from the rear,will then precess to the left. The yaw stiffness causes a nose-down pitching, which is in turn resisted structurally, resulting in thepropeller disc yawing to the right. The yaw stiffness on that side then causes a nose-up pitch to complete the cycle. This effecthas been termed the whirl mode, and its direction of rotation is opposite to that of the propeller. Violent Whirl Mode In a healthy airplane, the whirl mode can operate only withinthe limit of flexibility of the engine mount, and is damped out very quickly. However, if some structural element of the power-plant, mounting or nacelle is damaged—so that stiffness of the propeller supporting system is reduced below normal—then thewhirl mode would not damp out rapidly at high airplane speed. Powerplant installation damage in the Electra did not significantlychange the conditions under which the whirl mode might have been initiated, but it made that phenomenon, which is nothazardous in itself, become threatening in three ways. First, the greater flexibility of a weakened installation allowed the whirlmode more freedom and it could become more violent. Stiffness normally increases rapidly as deflections approach the mountlimits, but in a damaged installation this characteristic might be altered. Second, in an installation where the strength level isreduced, this, in combination with the increased force and violence of the whirl mode, could lead to further progressive damage andthus to further reductions in stiffness. Third, and most important, this changing spring constant is a condition wherein the frequencyof the whirl mode in an over-flexible structure could have reduced from its natural value (which is safely above the fundamentalnatural vibration frequencies of the wing), to lower values which approached the wing's natural frequencies. As the whirl modeprogresses, the frequency of the mode could approach the natural frequency of the wing, which would in turn tend to perpetuatethe whirl mode. Oscillations would then be coupled at the same frequency. This is a form of induced flutter, forced by a powerfulharmonic oscillation to occur at a lower airspeed than that at which classic flutter can develop. This could produce a catastrophicfailure through severe oscillatory divergence. Identification of the whirl mode phenomenon was the key tothe mystery which had baffled the aviation industry for months. The correction of the various wing, powerplant and nacelle defi-ciencies was, in retrospect, relatively simple once the problem was identified. The modifications were (1) to add additional mountsto stabilize the propeller should any member fail, or should break- age occur between the gearbox and the power section; (2) tostrengthen the nacelle structure by reinforcements and diagonal braces; (3) strengthening the wing by increasing the thickness ofthree lower surface "planks" and one upper surface plank, reinforc- ing the front spar and the wing lower surface and strengtheningeighteen ribs in each wing by additional diagonal braces and reinforcements. Dramatic flight tests were completed during the latter part ofDecember on a modified Electra. The FAA required a test of failure of the torque shaft and the torque shaft housing. Thesewere, in fact, removed prior to flight. The airplane was dived at speeds up to 418 miles per hour in this weakened condition.Severe flight loads were imposed by the pilot to record strains on the critical parts of the structure. Instruments revealed com-plete absence of the whirl phenomenon or any other adverse condition.Under FAA control, Lockheed is modifying aircraft at a rate which will permit all to be completed by the middle of 1961.As each airplane is completed, its speed restrictions are auto- matically removed and it is expected that airlines will returnto the high speed schedules as soon as they have a substantial part of their fleet modified.
Sign up to
Flight Digital Magazine
Flight Print Magazine
Airline Business Magazine
E-newsletters
RSS
Events
