2024 Robonauts CAD and Code Release

FIRST General Forum
, , ,
1

The Robonauts are excited to release the CAD, Code, Technical Binder, and Avionics board files from our 2024 robot, Twister.

Twister-2024 CAD, Code, & Tech Binder

Twister was by far one of the most complicated robots we have ever built and we are happy to share the CAD and code and answer any questions about our season.

The CAD can be found on GrabCAD here

The STEP file of the robot and the robot code can be found on our website under the robots tab.

We have also prepared a technical binder that highlights the details of each subsystem of the robot that is avaliable below
2024 Twister Technical Binder (Final).pdf (7.9 MB)

2024 Avionics Boards

We have improved our custom avionics boards that we used in 2023 and the design files can be found on our website under the robots tab.

2024 Avionics Boards
2024 Avionics Boards1173Ɨ959 310 KB

IPDU

Same purpose as the previous year, the 2024 IPDU needed even more channels, while shrinking the overall footprint. The 2024 design utilized two PCBs that stacked on top of each other. The PCB on the top left is the bottom stack, which housed the switching elements, additional protections, and status LEDs. The top PCB is a simple breakout board that only has the output connectors. Splitting two the roles into separate PCBs allowed the team to do fast and cheap iterations on the connector numbers and placement. The stack on the top right shows the complete IPDU with the PCBs stacked on top of each other with itā€™s case.

CAN Node

Bottom Left is the CAN node. This PCB allowed the team to implement the star topology with a optional on-board termination. Changes from 2023 included 18 channels and switching to the 2 pin Molex SL connectors.

Molex SL RIO Hat

Bottom Right PCB is our roboRIO hat which converts most of the roboRIO pins to our standard Molex SL connections. This gave us the positive locking feature while preventing FOD from falling into the channels in the roboRIO. Unused connectors on the PCB were populated with empty housings to ā€œcapā€ them off.

Additional notes

Custom PCBs went through a process to make them more rugged. After soldering, the PCBs went through a ultrasonic cleaner filled with a 50/50 solution of IPA and Deionized water. Next, ICs were then staked with grey RTV 3145 MIL-A-46146 (as seen on the top left PCB). Lastly all PCBs were then coated with an acrylic conformal spray.

Fasteners without Loctite are also staked and torque striped to provide a visual confirmation throughout the life of the bolted joint interface. Lastly all fasteners pertaining to avionics also include a washer under the head. This is especially important for the WCP PowerPole Adapter board where the bolt head kept galling the surface of the PCBs

Details of these processes can be referenced by NASA Standard 8739.1B and 5020B.

We are excited to Kickoff the 2025 season tomorrow and wish everyone good luck in FIRST Reefscape!

39 Likes
2

Iā€™m curious about this, because twister seemed pretty simple when i first saw it (intakes, turret, shooter, arm, and climber). How do the Robonauts define complexity and minimize it?

Great robot as always (the energy chain is among the coolest things iā€™ve seen in first)!

3

Itā€™s always a great day when Robonauts releases CAD!

This is cool. I really love the fact that you all apply standard NASA processes and engineering best practices to FRC robots. There may be other methods that work well for teams, but itā€™s such a cool thing to expose the students to the actual industry processes and standards.

11 Likes
4

Always love to see the Robonauts! <3

If youā€™re able to answer Iā€™d love to know:

What is the actual purpose of creating the IPDU as opposed to using something like REVs MPM?

  • Is it simply to add the extra protections (RPP, ESD, Thermal, etc) on those channels?
  • Is the switchable channel feature actually used?
    • If yes, for what purpose, I canā€™t imagine you would be saving much power from doing so, not to mention having to program around intentionally powering down sensors.

Thanks yall!

5

Beautiful, as always. I especially appreciate the CAN node, and I hope more teams look into doing something similar.

Can you expand on what some of these terms mean from the binder?

  • ā€˜APS coverageā€™
  • ā€˜Localized ferry setpointsā€™
6

I would say there is a strong correlation between how painful it is to make the robot work as intended and complexity. Twister was a very painful robot to get to work consistently and many of us feel we never really reached the peak that the architecture was capable of. Many of the pain points were with the note path through the robot and specifically feeding the shooter in a way that gave us a consistent shot into both the speaker and the diverter for amp and trap shots no matter the orientation of the note through the turret. We were constantly making changes to the shooter feeder roller geometry and materials all the way through champs to keep increasing performance.

We often struggle to minimize complexity. With 80+ students its easy to keep adding mechanisms and functionality to the robot so that there is more for the students to design, build, and program. We often feel that we can make any robot concept work given enough time and effort but Twister was close to being a robot that was too much for the resources we have. It took a lot of long hours, late nights, and many re-designs to make the robot work as we originally intended and we donā€™t really want to have a season like that again in 2025.

All of those extra protections are features that we desire but the bigger reason is that we can use our own connector standard. We can have a dedicated connector for each device rather than branching off to multiple devices.

The switching feature mainly exists to allow us to power cycle devices during a match for troubleshooting. We had a rough 2022 season where we had ESD problems with Limelights that could have possibly been fixed during matches if we could power cycle the device.

Its also just good engineering that gives our electrical students exposure to designing and assembling their own PCBs.

The term ā€œAPS coverageā€ refers to the fact that we geared the Absolute Position Sensor on our turret down such that even over 420 degrees of travel of the turret the APS allowed us to know the absolute position at any point without having to use some sort of homing routine or start the turret in the same position every match.

ā€œLocalized Ferry Setpointsā€ refers to our ferry shots that we used to shoot notes from the midfield area to our partners near the speaker. This was our ā€œferrying strategyā€ that we often used. We used localization with a Limelight 3G and odometry to track where we were on the field at all times and shoot our ferry shots to the same point on the field no matter where on the field we currently are.

10 Likes