Blog Post #6: Constructing My Second Robot

 Since I had completed the design of the second robot, this week I worked on actually building it. I needed to order all of the parts made out of metal as soon as possible to have the best chance that those parts would arrive in time for me to finish and test them before the end of the project period. While the original estimated ship date was on the 25th, still as I am writing this on the 27th, the parts have not shipped and it looks like they can only come on the 29th at the earliest so I have about one day of cushion before the end of the project period which is unfortunate. 

While waiting on those parts, I started to 3D print all of the parts that I could. The side armor and the mounts for the spinner are made out of TPU. When printing parts out of TPU, managing to get rigidity in key places becomes an interesting challenge. The first iteration of my spinner's mounts was way too bendy. If the spinner were to deflect while running, it could severely damage the robot itself and/or cut its own wires. To increase the rigidity, I made a few changes. One of the things I noted was that the parts were all very rigid around the places that had a lot of wall layers. To increase the presence of wall layers in my print, the first thing I did was skeletonize the part. While you normally do this to decrease the weight of the part, All of the new surface areas caused the printer to add more material around the edges increasing the weight. After this, there were still a couple of places that flexed more than I would have liked. To add wall layers here, I added small holes and extruded them about 0.7 mm into the part to reinforce the problem areas. I also 3d printed copies of the parts that I would have in CF and metal so I could see how everything was going to come together.

Chassis mockup with printed parts

The next thing that I had to do was get the top and bottom plates out of Carbon Fiber. I originally thought that I would be able to machine the plates in the Fab Lab. But after I did some looking around online, I began to doubt myself. Many sources online were all talking about how the dust was really abrasive and dangerous or not talking about it at all and nobody provided any concrete solutions. I thought I could easily implement though. Of the CNCs in the fab lab that could even machine carbon fiber, one of them, the Carvey, is enclosed with no ventilation which would make dust management a nightmare. The ShopBot on the other hand did have a dust shield and I could attach the shop vac to it which was promising. After meeting with Mr. DiGiorgio in the fab lab, we decided that it would be fine to use the ShopBot as long as I vacuumed up very thoroughly afterward. Throughout the whole process, I wore an N95 mask. While that mask was enough for one time, if I were to do this again, I would probably invest in a reusable respirator and use gloves. While cutting the parts, we ran into a couple problems where the CNC wasn't cutting deep enough, but after a couple more passes deeper and deeper, both of the covers came out really well and far easier than I thought it would be. Because I had doubts about the fab lab though, I had requested a quote from an external company to get the parts if the fab lab didn't pan out. Instead of using SCS which would have cost me hundreds of dollars for CF, I requested a quote from a service called CNC Madness. The quote they gave me was $17.68 for both covers which is cheaper than what I paid for one sheet of carbon fiber so I wish I knew about this before and would recommend this service instead of doing it yourself. 

Cut covers

The next thing I needed to do was connect all of the electronics. When I purchased the motor controllers, I asked for them without the connectors that came on the kit. While easy to use, I was scared they could disconnect from the motor really easily which would be really bad during a fight. Instead, I opted to solder the connections on myself. I also had to solder wires to the switch. The way I decided to implement the switch was to solder the red (positive) wire from the female JST connector, which plugs into the battery, into one terminal and another red wire to the other which would connect to all of the other components. Another major thing I had to do was bind my radio to my transmitter which was much easier than I thought it would be. Then came another problem. I seemingly had no place for all of these electronics. I thought the layout worked on paper but I ended up with a spaghetti mess of wires. In order to solve this problem, I hope to compress some of the wires with zip ties and electrical tape, lose the lever clamps, solder all of the wires together, and assemble the robot again, this time focusing on the electronics to ensure they all fit together. 

All the electronics in the robot