Build Week One
Design Updates
Closing out week one of build season, we have started to commit to one of our early robot designs that prioritizes fast speaker and amp cycles while maintaining a low center of mass and having an adjustable shooter to extend our range as far as possible. We’re making good progress in our robot CAD and combing out issues before they crop during fabrication and drive testing.
Full robot V1 (minus shooter). Crescendo 2024 - 0000 Main > full robor > Ver. 1.14.2024 10:17
Drivetrain
Current drivetrain with transparent bumpers to showcase side-skirts. Crescendo 2024 - 0001 Drivetrain > Drivetrain > Ver. 1.13.2024 18:16
For our drivetrain, we have settled on a 30”L x 28”W box tube frame driven by REV 3 inch MAXSwerve modules. Our drivetrain has a high bumper to allow for under-the-bumper intaking, coupled with polycarbonate side-skirts to funnel notes into our intake and keep notes from getting caught under our robot. We decided to increase our drivetrain dimensions to 30”L x 28”W, up from 28”L x 26”W, to give ourselves more design freedom, allow for simpler subsystem packaging, and provide even greater stability while driving. Our main intake–shooter system and our climber take up all of the central space of our drivetrain, where we would normally put our electronics, so this year we are going to add on side panels to mount our electronics on. Our drivetrain is our subsystem closest to being finalized and we will hopefully have it built and wired by the end of week two.
Intake
UTB intake mounted within the drivetrain. Crescendo 2024 - 0002 Intake > Intake Main > Intake > Ver. 1.13.2024 15:20
Our under-the-bumper intake is the first stage of our game piece path. It’s capable of pulling notes off of the ground with two rollers, and pushing them up an HDPE ramp into our shooter with a third roller. This intake facilitates compact and low-to-the-ground ground pickup and is much simpler compared to bulky, pivoting over-the-bumper mechanisms. We are aware of the inherent disadvantages of working with game pieces under-the-bumper, such as smaller intake width/reach, egregious frame space usage, and poor driver usability, but hope to mitigate many of those issues with a simple intake extension mechanism and additional side panels to provide space for our electronics.
Intake Extension
OTB intake extension. Crescendo 2024 - 0002 Intake > Intake Main > Intake > Ver. 1.13.2024 15:20
Another drawback of under-the-bumper intakes is limited intake reach and width. We address this by adding a static over-the-bumper intake that drops down at the beginning of a match and stays down to vector game pieces into our chassis. We expect this game to be very fast, with the possibility of high speed collisions during cycles, so we’ve taken measures such as doubling the intake plates to mitigate the impacts of T-bone collisions, adding a box tube beater bar to mitigate the impact of head-on collisions, and using HDPE mounting plates to absorb shock at the mounting sites. We’ve heard that vectored intake wheels may be ineffective on notes, and we will test that within the next week. If they don’t work as well as we hoped we are considering utilizing a design similar to what 7407 described in their Robot Concept Update 1 video that could vector notes without vectored wheels.
Intake extension bottom-up view. Crescendo 2024 - 0002 Intake > Dead Crossbar > Dead Crossbar > Ver. 1.13.2024 15:20
Shooter
We have been hard at work iterating on our shooter to get as much effective distance as we can out of it. We originally planned on a dual horizontal flywheel design; however, after considering the packaging of the shooter and seeing many videos from other team’s tests, such as 95, 4481, and SteelCityRoboticsAlliance, demonstrating that vertical flywheel shooters can deliver impressive range, we decided to pursue a vertical flywheel setup.
One of the main issues with a vertical shooter design is that adding spin to stabilize notes is more difficult. We have a few different ideas on how to solve this problem. The first one is adding a grippy surface to one side of the shooter to cause more drag on that side and add spin (i.e. hop-up). We like this idea because it is extremely simple; however, we are unsure about its effectiveness. The other idea comprises powering the left and right sets of flywheels with separate motors so we can vary the amount of spin. We like the idea of being able to easily adjust note spin but this design would add a significant increase in complexity and weight.
Our plan going forwards is to prototype a vertical shooter with separately controlled flywheels and use the data from that testing to decide whether we add it to the robot.
Vertical shooter prototype jig. Crescendo 2024 - prototypes > shooter > vertical testbench > Ver. 1.12.2024 23:53
We have decided to stick with a pivoting shooter design because we really like the flexibility. A pivoting shooter will allow us to shoot at the speaker from wherever without compromising our amp scoring. We were originally considering driving the pivot with a chain but after more thought and mentor feedback decided to swap to belt-driven to avoid the backlash a long chain would suffer from.
“From the moment I understood the backlash of the chain, it disgusted me. I craved the workability and certainty of belts.” — Ari
“Two Krakens driving a relatively light arm through a belt? This is FRC programming on easy mode!” — Max
The last aspect of our shooter design that we changed since our last update was the way that we deliver each note into the flywheels.
Shooter v1 top view. Crescendo 2024 - 0003 Shooter > Shooter> Ver. 1.11.2024 20:18
Our original idea, pictured above, was to have two compliant wheels hold the note in the plate until we wanted to shoot and then we would spin these feed wheels to deliver the note into the shooter. However, after a little thought, we figured out that as each note traveled through the flywheels the feed wheels would cause drag on the note, reducing our range. We have come up with a few ideas to solve this problem. The one we believe most promising is to use smaller feed wheels at the very back of the note to pinch it in place and cause very little drag when shooting. Another idea we had was to use a solenoid to block the note’s path into the flywheels and use a second solenoid to push the note into the flywheels.
We intend to prioritize our shooter this year and plan to keep iterating on it throughout the build season. Our current goal is to be able to shoot into the speaker from the front of the stage but the dream would be to be able to shoot into the speaker from the side of the stage.
Climber
Climber in context of our robot and climber fully extended in climbing position (bumper touching truss). Crescendo 2024 - 0000 Main > full robor > Ver. 1.14.2024 10:17
We have seen several ideas tossed around for climber mechanisms, such as elevators, arms, and even some more wild designs like grapple guns or nested PVC pipes. However, we see no reason that a COTS telescoping climber would be ineffective this season. Our analysis is that climb success will primarily be determined by the method by which a robot interfaces with the chain. Our solution to this takes heavy inspiration from one of 4481’s climber prototypes, using a hook that locks onto a vertical link and is held in place by the next horizontal link.
— Ari | 3636 Design Lead, Adam | 3636 Design, Max | 3636 Programming Lead
