Group 075-10: Advanced Manufacturing

Week eight began the first week of testing with the carbon fiber and PEI in our assembled melting chamber (seen in Figures 1 and 2). We began regulating the temperature of the melting chamber at 216 degrees Celsius, increasing the temperature by 5 degrees every ten minutes. An hour into testing, the pellets of PEI still had not melted, despite our insulation and the thermocouple registering temperatures in the 240-255 degree range. We increased our temperature drastically to 380 degrees, but because of the arrangement of the metal pipes, it was impossible to determine if the PEI in the chamber was actually melting effectively. Figure 1: assembled melting chamber, uninsulated Figure 2: assembled melting chamber, uninsulated, top view Our method of insulation was to wrap vinyl-backed fibreglass sheets around the body of the melting chamber. According to the thermocouple, which was registering temperatures higher than the heater setting, the insulation was effective. Unfortu...

Testing using the ABS prototype has continued with good results. The data continues to align with the theory that suggests a relationship between coating diameter and nozzle diameter. Our data analysis has brought forth an issue with our testing apparatus that is currently being investigated. The paint we have been using is water based; this means that while the paint dries, the diameter shrinks due to water evaporation. The percent solid that remains is not 100%, as it would be if the strand were coated with melted thermoplastic. To compensate for this in our data, we must identify how much water is lost when the paint dries. The water loss will be measured by weighing a sample of paint right out of the container, and again after it has completely dried. Once this is complete and accounted for in our calculations, our data can be further analyzed. Our prediction is an even more accurate proof of concept. Earlier this week, our melting chamber design was finalized and sent to two...

Testing has continued using a new iteration of the prototype. The output die has been modified with a threaded hole that allows for different nozzles to be used. Our design must interface with current 3D printer nozzles to allow for seamless extrusion, so nozzles that were designed to be used in the hot end of the printer provided a useful starting platform. We have chosen to implement nozzle sizes of 1.0mm, 0.8mm, 0.6mm and 0.4mm. These are all standard nozzle sizes that can be used on most commercially available 3D printers and are able to accommodate a strand running through the middle.  The underlying theory suggests that modifying this parameter (exit diameter) will allow us to vary the diameter of the coating. So far, our data reflects this concept; the coating diameter decreases as nozzle diameter decreases. We have found that varying the speed that the strand is pulled through the pool of coating material has little effect on the coating diameter.  The second...

The coated mono-filament produced by the first prototype was initially measured using calipers, which provided a reasonably accurate measurement for this initial design phase. As the design iterations progress, a more precise measurement procedure must be implemented as a quality control measure. The desired outcome of this project includes the possibility of a variable diameter filament produced in steps of 0.1mm and deviations will likely cause problems during extrusion from the printer. To fulfill this demanding requirement, a light microscope connected to a computer with analytical software was used to measure the diameter of the coated strand. Figure 1 is a snapshot of Trial 6 (200ms delay), this is the type of image that the diameter measurements were taken from. The system is calibrated using a calibration slide that provides a conversion factor in pixels per mm. The measurement scheme is illustrated in Figure 2. A series of three measurements were taken approximately 0.1 mm ...