Engineering topics
Publications and source records attributed to Thompson, William C..
This report describes a simple model for the effect of domain-irrelevant contextual information on the probative value of expert witness testimony. The model confirms prior observations by Thompson that such information always decreases the probative value. In addition, the effect holds regardless of whether the individual evidence items are conditionally independent or not.
The investigation was made to study the rough-water landing characteristics of a Gemini type of spacecraft. The investigations were made with a 1/6-scale dynamic model in a simulated sea state 4 rough water. Parachute letdown landings were simulated with the model at various yaw angles and horizontal velocities. The vertical velocity and landing attitude remained constant. The range of maximum lateral and longitudinal acceleration was from about 3-1/2g to 16g while that for the maximum normal acceleration was from lg to 15g. The range of maximum angular acceleration was from about 0 to 190 radians per second(exp 2). The smoothest behavior and the lowest angular acceleration occurred at the 90 degree yaw angle. The normal acceleration was near minimum at this condition.
Investigations were made to study the water-landing and certain grounds-surface landing characteristics of a Gemini spacecraft model. The water landing experiments were made by simulating paraglider and parachute letdowns with two 1/6- scale model configurations. Parameters included various combinations of attitude, horizontal speed, vertical speed, and landing skids extended and retracted. Investigations were made in calm water and in waves. The paraglider landings at horizontal speeds of 63 feet per second (19.8 m/sec) which resulted in a noseover or tumbling shortly after initial water contact. The maximum longitudinal acceleration of the model in calm water was about 14g units, and the maximum angular acceleration was 66 radians per second squared. In the parachute landings with the heat shield forward, the model skidded along the water surface on the heat shield. Parachute landings with the small end forward resulted in behavior similar to that of the paraglider landings. The ground-surface landings were made with a 1/3-scale model by simulating a parachute letdown with braking rockets, which were fired prior to touchdown to dissipate vertical velocity. In these landings, control of timing and aligning the rockets on the model was very critical, and violent behavior resulted when either rocket alignment or timing was in error. In the landings that were correctly controlled, the model either remained upright or slowly rolled over on its side.
Brief dynamic-model tests have been made at the request of the Federal Aviation Agency to investigate the use of a shallow pond of water at the end of a runway as a means of arresting jet-transport aircraft when they are forced to abort on take-off or overrun on landing. Such a scheme is of particular interest for civil aircraft because it requires no modifications or attachments to the airplane and no mechanical devices in the arresting system. A modification of this scheme that uses a flexible plastic cover over the water surface has also been tested. The purpose of this paper is to present the results of a dynamic model investigation which would aid in determining whether the water-pond arresting system could be used as a means of arresting airplane overrun.
Various configurations and approach conditions were investigated in order to determine the ditching behavior and the best ditching procedure. When ditchings were made without the float bags, the model rolled over on its side; when ditchings were made with the float bags inflated, the model remained upright. Late-flare and early-flare ditchings gave the same general behavior. Slight damage to the bottom surface of the scale-strength fuselage resulted for all test conditions.
An investigation of a 1/12- scale dynamically similar model of the Douglas F3D-2 airplane was made in calm water to observe the ditching behavior and to determine the safest procedure for making an emergency water landing. Various conditions of damage were simulated to determine the behavior which probably would occur in a full-scale ditching. The behavior of the model was determined from motion-picture records, time- history acceleration records, and visual observations. It was concluded that the airplane should be ditched at a medium high attitude of about 8 degrees with the landing flaps down 40 degrees. In calm water the airplane will probably make a smooth run of about 550 feet and will have a maximum longitudinal deceleration of about 3g. The fuselage bottom will probably be damaged enough to allow the fuselage to fill with water very rapidly.
An investigation of a 1/14-scale dynamically similar model of a panto-base version of the Chase C-123 airplane was conducted to evaluate the hydrodynamic characteristics of the airplane. The resistance, longitudinal stability, and spray patterns during take-off and general behavior in calm- and rough-water landings were determined. Brief calm-water tests were made to compare the initial vertical impact accelerations of the model with and without hydro-skis. Take-off stability was satisfactory for calm-water operation. A ratio of gross load to maximum resistance of 3,6 was obtained. Heavy spray reached the propellers only during ski emergence. The landing behavior in calm water and in waves 3 feet by 150 feet (full scale) was satisfactory for a normal range of trim angles. Initial impacts in calmwater landings resulted in vertical accelerations of about 2 1/2 with the hydro-skis installed and about 4g with the hydro-skis removed,
An investigation of a 1/14 scale dynamically similar model of the XC-123 airplane was made in calm water to observe the ditching behavior and to determine the safest procedure for making an emergency water landing, The behavior of the model was determined from motion-picture records, time-history deceleration records, and visual observations, Various scale-strength simulations were made to determine if the fuselage would be damaged and, if so, to determine the extent and location, It was concluded that the airplane should be ditched at a nose-high attitude of about 140 with the landing flaps down 45 deg., The airplane will probably make a porpoising run of about 415 feet (full scale) and the maximum longitudinal decelerations will be about 1 g, The fuselage bottom probably will be damaged and the fuselage will fill with water and sink to the wing level.
A ditching investigation of a model of the Convair-Liner airplane was made to observe the behavior and determine the safest procedure for making an emergency water landing. The ditching model was designed and constructed by the National Advisory Committee for Aeronautics. Design information on the airplane was furnished by the Consolidated Vultee Aircraft Corporation. A three-view drawing of the airplane is shown. The investigation was made in calm water at the Langley tank no. 2 monorail.
An investigation of a 1/18-scale dynamically similar model of the North American B-45 airplane was made to observe the ditching behavior and determine the proper landing technique to be used in an emergency water landing. Various conditions of damage were simulated to determine the behavior which probably would occur in a full-scale ditching. The behavior of the model was determined from high-speed motion-picture records, time-history acceleration records, and visual observations. It was concluded that the airplane should be ditched at the maximum nose-high attitude with the landing flaps full down for minimum landing speed. During the ditching, the nose-wheel and bomb-bay doors probably will be torn away and the rear of the fuselage flooded. A violent dive will very likely occur. Longitudinal decelerations of approximately 5g and vertical accelerations of approximately -6g (including gravity) will be experienced near the pilots' compartment. Ditching braces installed in the bomb bay will tend to improve the behavior slightly but will be torn away along with the bomb-bay doors. A hydroflap installed ahead of the nose-wheel doors will eliminate the dive and failure of the nose-wheel doors, and substantially reduce the motions and accelerations.
Results of previous model ditching tests of the Lockheed Constellation airplane are reported. Further model tests have been made to determine the probable ditching characteristics and the proper ditching technique for the airplane with the Speedpak attached. This paper presents the results of these tests. Design information was furnished by the Lockheed Aircraft Corporation. A three-vies drawing of the airplane with the Speedpak attached is shown. The tests were made in calm water at the Langley tank no.2 monorail.
A supplementary investigation of a 1/18-scale dynamically similar model of the North American B-45 airplane was made in calm water to observe the ditching behavior and to determine the effect of further conditions of simulated damage. The behavior of the model was determined from motion-picture records, time-history deceleration records, and visual observations. It was concluded that, during a ditching, the nose-wheel doors, bomb-bay doors, under surface of the aft fuselage and lower part of the bomb-bay aft bulkhead will be torn away. When such damage occurs, the diving found in the previous tests will not occur and the airplane will porpoise during the first third of the landing run and then run smoothly. Because of the extensive damage which occurs, the fuselage aft of the pilots' compartment will fill with water very rapidly and is, therefore, an undesirable ditching station. No change in the previously recommended ditching attitude and flap setting (6 deg attitude, flaps down) is indicated by these supplementary tests.