No description available.

Active damage control is a new technique which mimics the self-repairing capability of biological beings and can be used to greatly increase the structural integrity and fatigue life. This thesis describes two approaches used in active damage control: direct stress cancellation and indirect stress cancellation. Direct stress cancellation is illustrated using an example of active crack damage control with embedded shape memory alloy actuators. Both experimental and theoretical investigations have been conducted to analyze the damage control mechanism. The approach of indirect stress cancellation is demonstrated with active fatigue damage control of adhesive joints. The damage and active control mechanisms of the adhesive joints have been experimentally and theoretically investigated. / Master of Science

The interference-fit joint is widely used in machine design to connect two cylindrical machine elements. Its popularity comes from the simplicity of the assembly and the low cost of the production process. Even so, no study has examined the boundary non-linearities in the assembly of these joints. Moreover, the contact stresses and the stress concentration factors in interference-fit joints under torsional loads have not yet been examined in detail. In addition, the interface stresses and stress concentration factors in interference-fit joints subjected to bending loads have not been approached theoretically before. The sugar-mill roller is one of the oldest and most important examples of the interference-fit joints. The frequent fatigue failure of the sugar-mill rollers under reversed bending causes costly emergency line-shut downs. The versatility of the finite element method and the capability of the point-matching technique in handling contact problems have been combined, in the present work, to produce a surface-matching technique. It has been found that the complete-cohesion contact assumption may be acceptable for coefficients of friction ≥ 0.2, which is the case for most of the normally machined surfaces. An approach combining the semi-inverse displacement finite element method and the surface-matching technique has been developed to perform the torsional analysis. It has been found that the governing non-dimensional quantities are the ratio of the cohesion-length to the interface-length and the coefficient of friction divided by the load-level. The load-level is the ratio between the angle of twist of the shaft per unit length when the torque acts on the shaft alone and the amount of the diametral interference divided by the shaft diameter. A finite element model, using 8-noded solid elements together with linear interface elements, has been employed to locate the cohesion and slippage-zones in the sugar-mill roller under bending. The contact pressure increases along the compression side and decreases along the tension side. On the contrary, the interface shear decreases along the compression side due to bending and increases along the tension side. Subsequently, a larger slippage-length has been detected along the tension side.

The goal of this academic project was to study the effects of different variables on the damage initiation and progression around four bolt holes of a joint in carbon fiber/graphite epoxy composite coupons. The tracked variables included the type of layup, R values, stress levels, and damage mechanisms observed in each specimen. In-situ x-ray of the individual coupons recorded the extent of damage, mostly longitudinal splitting and bearing failure, as a function of the cycle count. The following lay-ups were included: [45/90/-45/02/45/02/-45/0]s, [04/45/03/90/0]s, [±5/65/(±5)2/-65/±5]s, and [±5/65/(±5)2/-65/5/65]s, In particular, the objective was to determine the stress levels at which detectable damage starts developing by applying 50,000 cycles at incremental stress levels. Once damage was initially detected, we typically raised the stress level 2.5 ksi and cycled another 50,000 cycles until damage reached a point where the bolt holes had elongated 10% of the original diameter of 0.25 inches. This type of testing was be continued for several different R ratios and comparisons were be made between the performances of different lay-ups at varying load levels. A finite element model was created in ABAQUS to help understand the stress fields within the laminates.