Browsing by Author "Dr. Jeffrey W. Eischen, Committee Member"

Now showing 1 - 6 of 6

Analysis of Low Velocity Impact on Pultruded Fiber Reinforced Polymer Plates
(2002-05-27) Motley, David Richard; Dr. Eric C. Klang, Committee Chair; Dr. Jeffrey W. Eischen, Committee Member; Dr. Larry M. Silverberg, Committee Member
Impact is one of the greatest design limitations involved in designing new composite products. The purpose of this research was to gain an initial understanding of the impact behavior of fiberglass reinforced pultruded laminates with vinylester and isopolyester resins. The effect of adding a protective layer of rubber to the laminates was also investigated. Several drop tower impact test were performed with a relatively high mass (12 lbs.) and a low velocity (80-150 in/s). These impacts produced severe damage in some of the laminates with only one impact. The 3D stitched laminates and the Duraspan™ laminates, which are used in Martin Marietta Composite's bridge decks, showed the least amount of visible damage of the laminates tested. The Duraspan™ laminates however, had two interlaminar delamination regions while the others only had one. This delamination was seen with C-san images of the impacted laminates. This extra delamination region did not decrease the tensile strength at all. Tensile tests were performed with a 1 in. strip from the center of the impacted samples in order to test the residual tensile strength. Finite element models were created with ANSYS/ls-Dyna nonlinear finite element software. These models were used to simulate the drop tower tests and then were extended to thicker laminates as well as different impact speed and impactor mass. Several models were also created to predict the effects of the rubber protective layer. These models were able to predict approximate stresses and strains induced in the laminates during the impact which were compared to the damage from the drop tower tests. The models predicted that the rubber layer decreased the stress and strain in the laminate up to 50%. The drop tower tests confirmed that the rubber aided the impact resistance significantly.
Atomistic Simulations of Fracture of 2d Graphene Systems and the Elastic Properties of Carbon Nanotubes
(2004-11-15) Jin, Yun; Dr. Kara Peters, Committee Member; Dr. Eric C. Klang, Committee Member; Dr. Jeffrey W. Eischen, Committee Member; Dr. Fuh-Gwo Yuan, Committee Chair
The essential feature of all the macroscopic mechanical properties of a material are governed by constitutive atoms and the basic laws of physics. The traditional descriptions of fracture phenomena in solids have been developed almost exclusively from continuum concepts. These continuum approaches have successfully described many fracture mechanism in solid materials and continue to be of great use. Nevertheless, they still have some shortcomings, such as the spurious singularity of the stress field at crack tip. Hence, atomistic predictions may open new avenues in the studies of microscopic origins of material failure behavior. Also the highly active researches on the lightweight nano-structured materials in the past decade revealed the emergence of interests in predicting the properties of materials in the atomic level before they are synthesized. In this dissertation the fracture mechanism of a nanostructure material has been investigated by atomistic simulations. Macroscopic fracture parameters have been examined from both atomistic simulation and continuum models. There is a very good agreement between atomistic simulation and theoretical results from LEFM for the energy release rate in a semi-infinite graphene sheet containing cracks. The atomic description of the stress field in this case is also obtained and matches very well with the linear elastic solutions. Then another case in which a center-cracked graphene sheet with finite width is proposed. The energy release rates are obtained in both global energy approach and local force approach. These simulations show that, in macroscopic fracture mechanics under small deformation, linear elastic fracture mechanics is sufficient for the description of cracking behavior for this covalently bonded material. The results merge the discrete (atomistic) and continuum (macroscopic) description of facture. Meanwhile, the method to calculate J integral in the atomic system is successfully developed. The numerical results of J integral agree very well with the energy release rate in the same systems. Then after a necessary modification on the Tersoff-Brenner potential, the critical value of J integral, denoted by , is eventually reached as the measure of the fracture toughness of graphene sheet. The mechanical properties of single-walled carbon nanotube (SWNT) have also been evaluated in this dissertation. Several elastic moduli of SWNTs using the MD simulations were obtained at the atomic scale. The values of the in-plane Young¬°¬Øs modulus, rotational shear modulus, and in-plane Poisson's ratio are in the range of existing theoretical and experimental results. It has been shown from simulations that overall the elastic constants of SWNTs are insensitive to the morphology pattern such as nanotube radius and thus the effect of curvature on the elastic constants can be neglected. Assumption of the transversely isotropic properties on the cylinder surface of the single-walled nanotube was confirmed by numerical calculations. Besides the conventional energy approach, a new method, which denoted as force approach in the dissertation, has been developed to analysis the elastic properties of carbon nanotubes. The results from two approaches matched very well. The advantage of the force approach is that it can provide more accurate prediction than the energy approach. Furthermore, the force approach can predict the nonlinear behavior without assumption of assumed total potential energy in quadratic form described for small-strain deformation in the energy approach.
Damage Localization From Sensor and Actuator Data
(2004-01-27) Camerino, Michael Joseph; Dr. Kara J. Peters, Committee Chair; Dr. Jeffrey W. Eischen, Committee Member; Dr. F. G. Yuan, Committee Member
Integrated sensor systems are becoming more prevalent in structures for the purpose of structural health monitoring (SHM). A limiting factor for many current SHM methods is that in order to locate damage, modeling of the structural response is required. The structural model can introduce significant errors, in addition to those in the sensor data, both through limitations in the mechanical models and manufacturing variations. This work presents a method of localizing damage that eliminates the requirement for an independent structural model. The method is based on generating flexibility parameters of the structure in pre- and post- damage states. In order to determine these parameters, a technique of loading the structure at each sensor location with actuators and subsequently measuring the displacement at all sensors has been developed. This allows the required information to be constructed from only sensor and actuator data. From the sensor data, a set of damage location vectors are determined. These vectors allow one to localize damage in one of two ways. The first analysis reapplies each set of damage location vectors as applied forces to the structure and locates the lowest regions of stress. The second approach, more applicable to real-time health monitoring, locates the lowest values of the damage location vectors. Both techniques have the ability to locate progressive damage. Simulations on a plate structure are performed for two sensor meshes (eight and thirty-two sensor locations). The results demonstrate excellent damage localization, and some indication of damage severity. Finally, an experimental demonstration of the method utilizing eight sensors surface mounted to an aluminum plate is presented for four applied damage cases.
The Effects of Batting Materials on the Performance of Turnout Thermal Liners
(2007-05-04) Heniford, Ryan C; Dr. Jeffrey W. Eischen, Committee Member; Dr. Roger L. Barker, Committee Chair; Dr. Behnam Pourdeyhimi, Committee Co-Chair; Dr. Timothy Clapp, Committee Member
The effects of fiber and constructional variables on the properties of hydroentangled nonwovens important to their performance when used as batting components in firefighter turnout systems are investigated. Para-aramid, meta-aramid and oxidized PAN constructions are characterized on the basis of thermal insulation, flexibility and durability performance. The contribution of batting properties to the thermal protective performance provided by multilayer turnout systems is examined for selected turnout lay ups. Optimized thermal liner systems are suggested based on layered constructions including the properties of the face cloth components.
Elastic Wave Propagation in Composites and Least-squares Damage Localization Technique
(2004-07-29) Wang, Lei; Dr. Fuh-Gwo Yuan, Committee Chair; Dr. Kara Peters, Committee Member; Dr. Jeffrey W. Eischen, Committee Member
The main objective of Structural Health Monitoring (SHM) is to be able to continuously monitor and assess the status of the integrity of a structure or its components with a high level of confidence and reliability. In general, the common techniques employed in SHM for monitoring structures and detecting damages can be divided into two categories: (1) vibration-based approach and (2) wave-based approach. Since wave-based approach can provide better local health status information and has higher sensitivity to damages than vibration-based approach, this thesis focuses on damage localization of plate structure using wave-based approach by first characterizing elastic waves in composite laminates; then using a time-frequency signal processing technique to analyze dispersive stress waves; and lastly a least-squares technique is proposed for damage localization. Exact solutions of dispersive relations in a composite lamina and composite laminate are first deduced from three-dimensional (3-D) elasticity theory. The dispersion relations containing infinite number of symmetric and antisymmetric wave modes are numerically solved. Then, to make dispersive wave solutions tractable in composites, a higher-order plate theory is proposed. The dispersion relations of three antisymmetric wave modes and five symmetric wave modes can be analytically determined. The dispersion curves of phase velocity and group velocity are obtained from the two theories. From the results of the 3-D elasticity theory and higher-order plate theory, it can be seen from dispersion curves that the higher-order plate theory gives a good agreement in comparison with those obtained from 3-D elasticity theory in the relatively high frequency range; and especially for the lowest symmetric and antisymmetric modes, dispersion relation curves obtained from the two theories match very well. In the Chapter of time-frequency analysis of dispersive waves, a Wavelet Transform (WT) is directly performed on a transient dispersive wave to extract the time-frequency information of transient waves. Consequently, the dispersion relations of group velocity and phase velocity can be mathematically obtained. Experiments are set up to verify the proposed WT method, in which a lead break is used as a simulated acoustic emission source on the surface of an aluminum plate. The dispersion curves of both phase and group velocities of the lowest flexural wave mode obtained from the experiments by using WT show good agreement with theoretical prediction values. Having group velocities verified from the experiments, a least-squares method is proposed to SHM field for iteratively searching damage location based on elastic wave energy measurements. The method is suitable for achieving automated SHM system since the proposed method is based on active damage detection technique and deals with the entire sensor data in the least-squares algorithm without the need of ambiguously measuring the time-of-flights. The simulated data are obtained from finite difference method in conjunction with Mindlin plate theory. Simulated examples for damage detection are demonstrated by using the least-squares method. Moreover, an active SHM system is set up to validate the feasibility of the least-squares damage localization technique. From the simulated and experimental results, it is shown that the estimated damage position by least-squares method gives good agreement with the targeted damage location.
Metrology Artifact Design
(2005-11-22) Folkert, Karalyn Faith; Dr. Ronald O. Scattergood, Committee Member; Dr. Jeffrey W. Eischen, Committee Member; Dr. Thomas A. Dow, Committee Chair
Part acceptance is based on dimensional inspection by comparison to the tolerance specifications of the part drawing. These measurements are often taken on Coordinate Measuring Machines (CMMs); but the dynamics of the machine will influence the overall measurement. Traditionally, a calibration artifact determines the static influences of the machine such as machine geometry. The goal of this project is to design and fabricate a calibration artifact that will test a CMM both statically and dynamically and determine the effects of those influences. The artifact developed is a ring gauge (6" ID, 8" OD, 1" thickness) that represents the typical size of parts of interest. The ring is 17-4PH stainless steel with a plated layer of electroless nickel (150µm) on the surface. Each face of the ring was diamond turned to a mean surface finish of 37nm RMS due to tool damage and machine error. Sine wave features were machined on the inside and outside diameter (ID and OD) of the ring using a fast tool servo. The wavelength of the sine wave varies continuously along the surface from long wavelength to short wavelength and back to the starting long wavelength over one revolution of the ring. The actual spatial wavelength of the wave sequences from 0.4mm to 6.4mm. The machined peak-to-valley (PV) is within 4% of the desired PV of 10µm. A reference flat surface was also machined on the ID and OD for static measurements. The final artifact is to be measured using a similar measurement strategy as a part to be measured; the same probe diameter is used with various measurement speeds in one orientation. The transfer function of the CMM dynamics is found by comparing the accepted swept sine wave with the measured wave. Each different speed defines a frequency range of the transfer function. By using this artifact to define the magnitude and phase characteristics of the dynamic system, the operator can make decisions referring to the machine's capabilities that exhibits an anticipated error and uncertainty in a measurement.