This week I have focussed on going forward with my project, planning further and continuing prototyping. I initially had the problem of trying to find a way of joining the two parts of the leg while allowing for them to spin independently, which I solved by drilling a hole through the centre of both pieces’ joint section, and attaching one part to a bolt which would sit in the other part, as to be an axel joined to one of the parts. I was then able to put the GT2 belt pulley on the independent piece so it can rotate it. This posed two problems; the initial issue was in attaching the pulley in a way which was strong, but temporary (as it is just the prototype), which was easily solved by substituting glue with blu-tack. The second problem was keeping the nut on the end on while the knee spun, which was solved by separating the pulley and nut with a washer. Although this is the current solution, it still slowly unscrews the nut so a new solution will be made later.
Upon joining the two parts of the leg, a new problem arose – belt length changes depending on the angle of the leg. This problem comes from the fact that one end of the pulley is on the leg, which means it moves when the leg moves, but the other end is attached independently to the chassis, and stays static throughout movement. This means that the lower pulley changes position relative to the upper one, and so the required belt length changes depending on whether it is closer or further. While a solution to this hasn’t yet been implemented, the current idea is to go for a redesign of the upper half of the leg, adding an extension out to the side for the upper pulley to sit, so it moves with the rest of the leg. This solution has a problem in its own, although more annoying than concerning. This solution would require a thicker leg, as the motor is quite tall and the two pulleys have to be in alignment for the belt to work. This can be done easily, however will increase both printing time and material.
Other than these problems, the project has been developing smoothly, and I have been able to solve the more important problems as they come. Although my calculations for most the problems aren’t entirely reliable, they are enough so that I can continue with the prototyping phase. I believe this project is on track to being completely developed by the due date, although I am not as confident with having properly coded it by then.