Continuous fiber ceramic matrix composites (CFCCs) are currently being developed for a variety of high-temperature applications, including use in advanced heat engines. For such composites, knowledge of porosity distribution and presence of defects is important for optimizing mechanical and thermal behavior of the components. The assessment of porosity and its distribution is also necessary during composite processing to ensure component unifo4ty. To determine the thermal properties of CFCC materials, and particularly for detecting defects and nonuniformities, we have developed an infrared thermal imaging method to provide a "single-shot'' full-field measurement of thermal diffusivity distributions in large components. This method requires that the back surface of a specimen receives a thermal pulse of short duration and that the temperature of the front surface is monitored as a function of time. The system has been used to measure thermal diffusivities of several CFCC materials with known porosity or density values, including SYT.,RAMICm SiC/SiNC composite samples from Dow Corning and SiUSiC and enhanced SiC/SiC samples from DuPont Lanxide Composites, to determine the relationship of thermal diffusivity to component porosity or density.

No abstract

An analysis of the flow of a second‐order fluid is presented. Reference values for some variables are defined, and with these a non‐dimensional formulation of the governing equations. From this formulation, three dimensionless numbers appear; one is the Reynolds number, and two numbers that are called the first‐ and second‐dimensionless normal stress (NSD) coefficients. The equations of motion are solved by a finite element method using a commercially available program (Fidap), and the steady state converged solution was used to measure the die swell. The factors that influence die swell and that are studied in this work include: the die geometry for circular cross sectional dies, including tubular, converging, diverging, half‐converging/half‐tubular shapes; fluid characteristics such as Reynolds number and first‐ and second‐DNS coefficients (both positive and negative values); and flow rates, as determined by the maximum velocity in a parabolic velocity profile at the entrance to the die. The results suggest that shear and deformation histories of the fluid directly influence not only swell characteristics, but also convergence characteristics of the numerical simulation. © 1999 John Wiley & Sons, Ltd.

No abstract

No abstract