this week our assignment was to build a machine as a group that could had a mechanism, actuation, automation, application and we were supposed to build the mechanical parts and operate it. it is definitely the most exciting week and really fun one because we all get to work on many pieces all together.

we started by brainstorming for two nights in a row (wednesday + thursday).

we had many cool ideas (some much harder than others) but we decided that we wanted to prioritize projects that could be easily divided into parts, took good advantage of the materials supplied (which included aluminum extrusion, motors, motor controllers, belts, and joinery of many types), the complete list of things is here.

friday: was our first night of work and quentin, jake, claire, and leo were all there to help which made the energy fun even though i was super tired. we got started by diagramming and understanding corexy, our choice for cnc kinematics for speed and structure. the two main advantages of the corexy are that the motors can stay in place (they don't need to move with the cart), and they're very fast.

we also made some cute feet using the flexible resin in the sla printers (with spikes) so that the machine had a sturdy and sticky surface to stand.

we started by playing around with the material, setting a workarea, and designing some 3d printed parts (together with zach and umut).

saturday: we assembled the machine but there were some obvious design problems. the main ones were that the cart and casing were too loose on the rails and so we would never get precise movement (it would just slide around sloppily), the idlers weren't totally in plane, and biggest of all is that our designs really pared down the machineable area (EXTREMELY).

sunday: you can see that we almost had halved our workable space... uh oh. so we redesigned, and quincy took over this iteration as he had already built out the whole machine in solidworks. we had a team brainstorming session and then printed out the new parts.

monday:then we had a whole issue because we didn't have enough 6mm belt. quentin and jake brought some over from cba but it turned out that it was 9mm so i went home thinking i had some but it was also 9mm. we ended up rearranging the belt and we didn't have quite enough no matter how we tried, so we ended up moving the y-axis a little bit closer which took a few cm off of our workable area. it ended up being totally fine and allowed us to cut the belt in half and attach the second motor.

we flashed the motors with the modular things firmware, and connected the motors and used the createSynchronizer function to start playing around with controls.

we had a few adjustments to make, including adding some washers to level the plane a little bit and working on tightening screws just enough for good pressure but not enough to bind the machine, but it was working really pretty well. all credit to quincy for the designs!

we put some paper underneath and a pen and paper so that we could start testing the kinematics and understanding the controls. drawing the first line was a BIG DEAL, and then we started to understand the dynamics and drew a circle. this also felt like a big deal.

up till now will and i had been running the controls together, but then people headed home and i took over transforming the code so that we could actually take input arrays and draw what we wanted! i worked together with catherine to make sure that we were transforming images into arrays of arrays that contained x and y coordinates as well as a 0-1 value that indicated flow for our pump. catherines code transformed images into points, and then my code moves the machine.

and we tested drawing a smiley face with inky-water! and it worked! big excitement.