AP projects 2015
Posts tagged CNC milling
Today as a part of our minor our student group was introduced to CNC milling and 3D printing with Ultimaker2. Since we were free to choose the tryout objects, we decided on one of the .stl files we extracted from the CT scans.
“Harry”, as our focus object, seemed like a great begin. Yet just after loading the files to Cura (the Ultimaker software), we noticed that the model was not completely straight. That was due to its original position during the scan. In other words, the object nested in-between two other objects was not completely parallel to the ground plane. This we tried to fix by manually rotation in Cura. In the end, it was not perfect and due to that the bottom edge of the cup was sketchy. Moreover, Ultimaker 2 seemed not like the right machinery for such task. That was mostly because we had many open, hanging edges and pieces which had to be supported. The latter is possible with the same material, but that leaves clear marks on the surface.
Parallel to 3D printing we also did some CNC milling. Just like in Ultimaker, we used “Harry”. For this production technique we used DeskProto. This program both translates the files for the machine and helps to create a frame for the object (mostly necessary to get a clear reference point). However, the latter can also be manually done in other CAD software.
Production of the object took merely 15 minutes, but we did not strive for the highest resolution. That meant that we took the biggest cutter available (d8mm) and got a sketchy cup with clear stepping. Moreover, due to the cavity in the cup and flexibility of the material, mistakes were made (seen in the picture below) and the model was very flaky.
Since we had more than 1,5h left, we proposed to make another model in CNC machine. This time to see how much detail we could achieve. For this task we chose “Hermione” as the model, yet due to the time limitations we could only take a piece of it. We worked on the object from 3 sides, starting with 8mm cutter and finishing with 4mm. This procedure took at least twice as long as the previous one. What is more, during the first try the foam melted, completely destroying the model. In the end, results of the second one were not as clear as in the digital model, but still quite amazing: knowing that we used very soft foam, could not precisely put the model on the reference point and that we did not use the smallest cutter.
In conclusion, the CNC milling could be an option for the final product, if we went for a single material transparent/translucent look. This would be possible by milling stacked and glued plexiglass. 3D printing in single material is also very interesting, but with this we would have to sacrifice some of the qualities of the end product (historical footprint, practicality, aesthetics).