Mahdi Ashrafi

Effect of processing variables and fiber reinforcement on the mechanical properties of wood plastic composites

Wood plastic composite (WPC) specimens were fabricated in shapes relevant to flute (musical instrument) production using African blackwood powder and phenolic resin in a hot compression molding setup. The roles of composition and processing parameters on the mechanical properties (flexural strength, elastic modulus, and impact failure energy) were systematically investigated. Cracks were observed in composites with more than 70% wood particles due to the formation of gas in the system during manufacturing, and lack of fluidity in the system to flush out entrapped air and the produced gases. Based on these results, the optimum temperature, pressure, and wood volume fraction for manufacturing WPC in the form of a flute is developed. A further series of experimental procedures were performed to improve the mechanical properties of WPC samples by studying the addition of short glass fibers to the molding compound prior to hot pressing. The results showed that the addition of short fiber did not improve the strength of WPC but rather reduced its strength compared to unreinforced composite. This was attributed to lack of bonding between the short fibers and composite matrix.

High Strain Gradient Measurements in Notched Laminated Composite Panels by Digital Image Correlation

Digital Image Correlation (DIC) was used to measure the in-plane strains induced in laminated graphite-epoxy panels containing a central circular hole and loaded in tension. Panels with three different stacking sequences and two hole diameters (0.25 and 0.375 in.) were tested. Measured strains were compared with predictions based on the Savin solution. Reasonable agreement between predicted and measured whole-field strain patterns was achieved. However, it was found that filtering of the DIC strain data, which is performed routinely to reduce noise, can mask the high strain gradients that actually exist near the edge of the hole. It will be shown that stacking sequence of the composite laminate as well as the hole size in addition to the correlation variables such as subset, step, and filter size have direct effect on strain values measured by DIC. In this study, strain measurements from the predictions were used to tune the correlation variables.

Measurement of Strain Gradients Using Digital Image Correlation by Applying Printed-Speckle Patterns

Digital image correlation (DIC) measurements were obtained using speckle patterns produced using two different methods. In the first case, speckles were produced on the surface of a test specimen using white and black spray paint, which is probably the most common approach. In the second case, a computer code was written that creates a random-speckle pattern digitally. The digital-speckle pattern was then printed on white paper using a standard laser printer. The speckled paper was subsequently bonded to the surface of a test specimen using a commercial strain gage adhesive. Strains were measured using both types of speckle patterns for a thin plate with hole and for a U-shaped beam loaded in tension. Strain fields measured using the painted and digital-speckle patterns are essentially identical and are well predicted by the corresponding elasticity solutions.

Embedded resistive heating in composite scarf repairs

Composite scarf repairs were cured using heat generated by passing an electrical current through a woven graphite-epoxy prepreg embedded in the bondline. Resistance heating using the embedded prepreg resulted in a more uniform temperature distribution in the bondline while preventing any potential thermal damage to the surface of the scarf repairs. In contrast, conventional surface heating methods such as heat blankets or heat lamps lead to large through thickness thermal gradient that causes non-uniform temperature in the bondline and overheating the outer surface adjacent to the heater. Composite scarf repair specimens were created using the proposed embedded heating approach and through the use of a heat blanket for circular and rectangular scarf configurations. Tensile tests were performed for rectangular scarf specimens, and it was shown that the bond strengths of all specimens were found to be comparable. The proposed embedded curing technique results in bond strengths that equal or exceed those achieved with external heating and avoids overheating the surface of the scarf repairs.