I made this thread after realizing that our team didn’t have anything posted in the Robot Showcase sub-forum.
Here is a link to the original discussion and picture.
Feel free to leave any questions or comments in either thread.
How reliable do y’all think the side ramps can be? I’m still debating if I trust climbers that lift another robot by just supporting the underside of them, but they seem more and more prevalent.
I think the main issue is going to be lining up right when driving onto the ramps. If a team drove up onto our ramp but was off to one side or only partially on, then there would be a greater chance of falling off as our robot slowly climbed upwards. One of the reasons we decided to go with two carbon fiber “forks” with aluminium bracing was because a robot would get “stuck” as soon as their wheels fell through the holes, making it harder for it to fall of the side or front.
Another smaller issue that I think our team has managed to fix very well is only supporting one robot on one side. Done improperly, there is a chance that the main robot could twist and bend, also causing the robot on the ramp to fall. Our team has made steps to prevent this, by using two hooks that attach on each side of the rung, bracing the elevator with carbon fiber to prevent twisting, and using a climbing assist which pushes against the insets on the scale on the sides of the rung.
In short, I think that if a team can do ramps properly, the pressure falls less on the teams actually climbing and more on the teams coming along for the ride, wondering if they can actually center themselves properly when the time comes.
Disclaimer: I am not one of the team engineers, simply the guy watching them 24/7(1).
At the end of our elevator gearbox is a disk brake and a disk brake rotor, similar to one used on a bicycle. Ours is pneumatically controlled. Whenever the elevator is told to not move up or down via code, the brake is actuated, as shown in this slightly out of focus shot taken during reveal filming. When the brake is active, the motors are unable to spin, making the whole thing stay still.
You can see the custom gearbox and attached disc brake for a second at 0:45. We mounted it on the on the same shaft as first stage of the gearbox after the motors so it has way more mechanical advantage than it needs, but certainly gets the job done.
If you place 3 cubes high on the outermost edge of the plate will 4 haphazardly placed cubes on the opposite plate get control back? 5? I’m sure you guys tested it. That’s a really great machine feature! Rest of the robot it solid as well. Awesome job!
Wow thanks, that would be super awesome! Maybe we’ll see you there
Question: what’s the purpose of the pneumatic cylinder on the arm? Why not just use springs or a gas shock?
On the carriage, there were two mechanisms that we thought could use a large constant pulling force. One was the counterbalance for the arm, and the other was the rope tensioner for the elevator. The second of which greatly benefited from being able to depressurize the system for maintenance. in the near future we will be looking at lighter weight solutions such a gas springs for those applications in order to improve the robot’s agility.
Thanks a bunch! That means a lot coming from someone with one of my favorite robot designs so far.
I’m not actually sure what the outcome of that would be. We’ve never really looked at making power cube towers (towers of power?) as a strategy. With that shot in the video, we wanted to demonstrate that we would likely run out of cubes before we would run out of reach. However, we don’t own enough cubes to fill the plate 3 layers tall so we made a stack.
Nice machine. What was the gear ratio for the wrist. Did you try only one bag motor and and later found out 2 was needed? Still not clear why the piston was needed. Can you clearify that.
Thank you! The gear ratio on the wrist is 438:1. We used two bag motors on it simply because we had used that motor setup on similar arms in the past with good success. After testing we decided that the arm had much more torque than necessary, and was slower than we wanted. So we have sense moved to a single 775pro with the same gear ratio. The change seems good so far, but we need more testing to be sure.
Wow, just wow. So far this appears to be a robot that will absolutely dominate the field. Your robot reminds me a lot of 118’s bot from this year, but simplified in to an elegant design. I love the implementation of the continuous elevator. My only question is why are you guys running only 3 775 pros per side and not 4 per side for the drive train.
Sorry I missed that question on the first pass. The piston on the wrist is used to counterbalance the weight of the arm so that it can stay upright in the starting configuration, and so that the performance of the arm is constant while going up and going down. To save weight in the upper part of the robot, this cylinder will likely be replaced with surgical tubing before our fist competition.
The other piston on the carriage is used to tension the elevator rope. It may also be replaced with surgical tubing at some point, but it is nice to be able to reduce tension on the system for maintenance by venting air pressure.
That wheel spins freely. When the corner of a cube hits the wheel, it is forced to rotate in one direction or the other as the front wheels pull it in. Once captured, the wheel also registers with the handle hole of the crate, to improve our hold on it.
Thank you! I hope it works well on the field at our first event next week. The drivetrain was originally designed with two CIM gearboxes, but we were unhappy with the performance for a number of reasons, so we made the switch to three 775pro gearboxes at the beginning of week six. because of the space we had available, a three motor gearbox was easier to fit. We had also previously decided that winning pushing matches was not a design criteria, so we felt that we could use amperage limits to protect the motors so having four motors to distribute the load was unnecessary.
That’s such a cool idea! Maintenance is a bit of a pain with our chain lift. I’m not sure if we could have done something similar with chain but that sounds quite convenient. Does that mean that your elevator is not tensioned when the robot is off, then?