Robot Rebuild

For this offseason, our plan is to rebuild an entire robot based on a new archetype that our team has never done before: an elevator. For the official season, our robot was a double jointed arm that had a shoulder and an elbow, but no wrist. Here were the problems with the robot:

High Center of Gravity:

As the driver of the robot, I found it really difficult to drive as aggressively as I wanted. This was because the robot had a high center of gravity. In one of our matches at our second competition, I drove off the charge station at a weird angle and caused the robot to tip. The robot had tipped around three to four times throughout the entire season. Even though this may sound like a little compared to other teams out there, it was enough to make me want to scrap the entire robot.

Slop:

Having built the arm from thunderhex shaft, there was major slop in the arm (around 15-20 degrees). This was more than enough to throw the motor encoders off. Having learned a lot about mechanical arms this year from the other teams that implemented them onto their own bot, the best way to have done this would have been to use a deadshaft and directly drive arms instead of doing a live axle and having the arm be driven by the shaft. Because thunderhex shafts are slightly rounded, when it's in the hub, it's not a tight fit. This results in slop, but having an arm so long, having little slop can result in the arm position to be majorly off. On a long arm, a few degrees of rotation results in a large travel distance despite the small rotation.

Belt slipping on elbow joint:

Having trouble with tensioning the chain for the shoulder joint, we decided in design that it would be best to drive the elbow with timing belts. Although this solution did work, it did not work well in specific scenarios. When driving the robot with the arm somewhat extended out, if the robot were to crash into a wall, the belts would slip on the pulley. This threw the elbow motor encoders off by a lot. Every time this would happen, our manipulator would have to manually reset the arm which cost us time during our match. In order to keep a low cycle time, we should not have to be doing this.

Despite having made poor design decisions during the season, looking back at the season as a whole, we did not do entirely bad. I would say our season was decent, although we fell very slightly short of qualifying from the World Championship (just needed one more playoff win at states). The robot performed decently well compared to most of the teams in the district, but I was still not happy with where it got us. This is why we will be building a new robot for our offseason event