FRC 3506 YETI Robotics | 2024 Open Alliance Build Thread

EHutson · November 10, 2023, 8:14pm

Hi Everyone! Welcome to the YETI Robotics 2023-2024 Open Alliance thread!

We’re super excited to be back for the 2024 season! We loved being a part of the Open Alliance community for the past three years and look forward to continuing to be a part of this community.

Let’s start with a little bit about our team. We are FRC team 3506 YETI Robotics based in Charlotte, North Carolina, established in 2010, with our rookie year being the 2011 season, growing from 17 to 70 students going into the 2024 season. Our team is also part of the Queen City Robotics Alliance (QCRA), which supports our area’s various FRC, FTC, and FLL teams. Our community team welcomes anyone interested, as everyone should be supported due to a potential lack of assets. During the offseason, we meet once a week, and during build season, we meet weekly from Wednesday - Friday, 6 pm - 9 pm, and Saturday from 9 am - 9 pm. We had just about our best season with Charged Up and are endeavoring to keep that going into 2024!

One new development for our team: our new meeting and building space! We have moved from an old warehouse to an industrial office space at 2102 Cambridge Beltway Dr, Charlotte, NC 28273, for anyone interested. We are working on an offseason recap that will be coming out soon and will give updates once a week during the preseason and about every other meeting during the build season. We will post progress videos during the build season on our YouTube channel and use GitHub to store our programming data. As a SOLIDWORKS team, we are migrating to 3DExperience for CAD data. We currently do not have that setup, but we will post updates as soon we get it situated.

Feel free to ask any questions and have a great day!

Resources:

YETI Robotics - Team Website

YETI Robotics - YouTube

YETI Robotics - GitHub

YETI Robotics - Instagram

YETI Robotics - Facebook

reeti.praj4 · December 3, 2023, 11:40pm

Hello everyone! Welcome back to another issue of the YETI Open Alliance blog! One of the biggest updates is that we have signed a contract to build a full field at our new facility, which is under construction. Photos of our new space are below.

The week before Thanksgiving, we did not have a general session, and each subteam did its own thing. Our Mechanical subteam listened to a presentation on basic machine design by one of our mentors. We covered subjects such as using roller chains and belts, gear sizing and gearbox design, etc. Our Controls subteam worked on creating a robot project from scratch, and our Marketing subteam worked on preparing the impact essay and a few other awards.

A major event that we just started is our annual CADathon, which we kicked off this past Thursday, as this is a fun and engaging way for students to begin using Solidworks outside of online tutorials. The week before Thanksgiving, we began to prep students for CADathon by making sure SolidWorks was installed for our members, among other things.

This year, we began to test 3D Experience, which we use to manage our CAD data. This CADathon is based on the 2006 FRC Challenge, Aim High, with all rules the same, except modern components are permitted. We have five teams, each with about ten people with skills in different subteams. The CADathon is a week-long competition where each CADathon team does a 15-minute presentation for the whole team at next week’s meeting.

That is all for now! Have a great day!!

David_IsMe · December 8, 2023, 2:40am

Thanks so much for having me out earlier this week to tour your new space best of luck at Asheville and Meck, See you at DCMP!

EHutson · December 8, 2023, 3:54pm

You are welcome and thank you for coming out! See you at DCMP!

reeti.praj4 · January 7, 2024, 10:25pm

Hi Everyone! We hope everyone had a great Kickoff!

This year, we hosted our Kickoff at the new location of the QCRA Zone with the privilege of having two other teams present: Team 4290 Bots on Wheels and Team 9005 Avian. After the game introduction, our team split into groups where most of the team analyzed the game and brainstormed ideas while leadership began developing a game strategy.

Below are prototype ideas that our team came up with, ranging from various intakes, claws, and shooters targeting the different missions of this year’s game. Most of our inspiration was from previous games, such as 2011 Logomotion 2013 Aerial Ascent.

|215x287.98489197440114

|247x352.9956897417205

We initially made a list of priorities during the leadership’s strategy meeting. We discussed what aspects of the game we would prioritize to gain ranking points while taking into consideration the types of robots we would expect to see during competition. We also discussed match strategy to strategize a way to optimize our scoring potential efficiently.

Prioritization List

Drive!

Higher Score/Melody RP Prioritization

Ground Intake
High shooter
Amp
Climbing
Trap
Buddy Climb

Stage/RP Prioritization

Climbing
Intaking/low scoring
Trap
Amp
Buddy Climb
High Shooter

Combined Priority List (top priorities)

Ground Intake
- Can score in Trap and Amp
High Shooter
- Variable Angle
- Shoots into Speaker only
- Pass-Through (intake a shooter on different sides)

*Climbing

Separate Climber
Be able to climb with one other robot
Score in Trap
Trap
- Score with intake
Amp
- Score with intake
Buddy Climb

Autonomous Priorities

Shoot Preload High and move (absolute minimum)
Ground Intake game pieces and shoot high (in addition to high preload shoot)

We’ll give updates in a few days and post archetype data and other visual specifics of our robot prototypes as soon as we finish with the CAD illustrations we’re working on.

We hope everyone has a great season!!

reeti.praj4 · January 13, 2024, 2:32pm

There has been much progress during this first week!

We’ve been focusing on developing a strategy for this year’s game and found some aspects challenging. Being able to score Trap means not having to do triple climbs and not relying on a human player to score the high note. We found that a reliable Trap mechanism will almost guarantee one RP during qualification matches, provided that another robot climbs to obtain harmony. Climbing is crucial since two robots are required for the Stage RP. This game will be defense-heavy, as there aren’t any protected field zones near the Speaker (not counting the podium); we concluded that swerve drive would be the best option to move swiftly and efficiently on the field. Going under the stage will allow us to avoid defense during matches; considering this, our robot has to be relatively small in size; this could also aid in general movement on the field since visibility through the two stages isn’t optimal. Efficient subsystem integration and pathing during this game is essential to successfully scoring during the autonomous period, which can assist us in achieving an RP.

Robot function list of priorities - final:

Drive
Ground intake
Shoot into Speaker
Climb
Score in Trap
Score in Amp
Intake and shooting from different sides

Our Mechanical Subteam has had many prototypes underway in CAD over the last few days. The CAD team decided on a robot archetype, which is a handoff design. This initial design consists of a roller bar intake that flips out and hands the game piece to the shooter. The shooter can tilt and extend to score the Trap and Amp. We have two telescoping climbers for the endgame, aiding the buddy climb. Much work has been put into the geometry and sizing to ensure all the subsystems can work together.

Here is a CAD illustration of the top-level handoff

And the roller-bar intake prototype

Along with our illustrations of our telescoping climbers, giving you a general understanding of how this would look like on the field.

Other members of the mechanical subteam are working on prototyping a shooter and intake. The shooter is a one-sided rail shooter, aiming to shoot accurately and effectively with a flywheel on one side, as we are aiming for a lightweight and simplistic shooter. The intake prototype is mainly to feel compression with the game piece. We are still working on setting a set compression ratio.

Our Controls subteam has been working on updating our RoboRios and other firmware for our robot. We have also been researching different options for subsystems on our robot. For example, for our flywheel, we’ve been exploring methods such as Bang-Bang, which is more simplistic in its coding and tuning, versus PID, which is generally more accurate but more complex. We’ve also made our code skeleton and intake code in preparation for this game. The link below takes you to our GitHub, giving you access to all our programming specifics for this season.

More to come. Have a nice one, good luck teams!!

EHutson · January 14, 2024, 8:17pm

Hey Guys! Evan here and one of YETI’s two CAD leads. I wanted to give a quick update on what we did yesterday CAD-wise. We meet 9-9 on Saturdays, so this is our big day design wise. We are currently trying to figure out how to make our CAD public; we will update you with the file very soon! If you have any feedback please let us know and we are more than happy to answer any questions!

For our intake, we are using an over-the-bumper design that is 14 wide. We are going with that width so we do not have to worry about vectoring the game piece; however, that very well might change in the future. Our intake is on a pivot using the Spline XL shaft, and rotation is by a single Kraken. We have a 1.33:1 ratio for our intake rotation and our building in a hard stop for the game piece. The WCP 1.125 in diameter rollers are what we are using for the intake itself and we currently have .2in of compression. We built the gearbox into the gussets to support the shaft for weight-saving and more effective packaging. The intake pivot is currently on a 60:1 ratio.

We have refined our Top Level, added more detail today, and changed a significant bit of geometry to ensure everything integrates well. For our climber, we are using the WCP Greyt telescoping climbers. We are cutting the tubing down as small as possible and using a 3-stage telescope. We currently have a 19 fps ratio on our swerve modules which are Swerve X.

Our Shooter uses a dual flywheel design with a 1:1 on a Kraken. The shooter is on a pivot to have an adjustable angle and to enable us to score in the trap and the amp. The flywheels are currently horizontal but we are exploring using two rollers on top of each other instead. That decision will come down to the testing that we will complete this year. We do have two vertical rollers on the intake side of our shooter to be able to help facilitate the handoff and hold a game piece. Our Shooter is located on top of an elevator which enables scoring in the trap and gives us a greater range of scoring capabilities by having more positions to shoot from. The shooter pivot is on a 128:1 ratio to make holding an angle easier and enable us to have really fine adjustments.

Thank you for reading!

EHutson · January 14, 2024, 8:58pm

So we can get our CAD out quickly we made an Onedrive link! This folder contains our top level and all its references. I will continue to update it and I will post when that has been done. Also please let me know if the link works. Thanks and have a great day!
https://1drv.ms/f/s!AtzPczL4ui-sgYcUOc51PB1Sijywwg?e=DSoiKp

reeti.praj4 · January 15, 2024, 8:19pm

Much testing went underway this past Saturday!

The majority of our field has been completed, allowing us to test the dynamics and constraints of our prototype. A shout-out to our mentors and parents!

The CAD team worked on finishing potential designs in CAD and played with the geometry of the subsystems to ensure that we could score Trap and Amp, the two main requirements for our robot. We also worked on putting the electrical components onto our robot to achieve a clean and organized finished look for wiring.

We altered our previous prototype of the one-sided rail shooter to have one flywheel per side. This allows for greater compression to develop a shooter that will be used to test vertical compression when shooting. This design will potentially go onto our robot if the tests show that it is reliable and better than our previous design.

When testing, we concluded that the trajectory of the note was lower. To improve this, we tightened the chain of the flywheels to increase the compression. This would heighten the note’s altitude in order to score the speaker efficiently. We had an initial compression of four inches, which we changed to six inches on the game piece. We found this to be more accurate for scoring.

The robot was initially 93 inches away from the speaker. We discovered that the farther away from the speaker, the less stable the note’s trajectory was. After moving the robot up to 91 inches away, we found the path smoother, leading to more accurate scoring and less frequent bounce-outs.

The dimensions of the blue plate on our shooter are 27.5 inches, and its width (motor to motor) is 29.75 inches.

Feel free to ask any questions. Have a nice day!

reeti.praj4 · January 21, 2024, 7:22am

Hi all! Here’s our recap of Week 2!

This week, our mechanical subteam focused on building and testing prototypes. We developed a new vertical shooter with three flywheels on the bottom and top hex rods. This shooter was very effective, and as soon as we saw its potential, we quickly began designing a more official prototype that could be mounted easily onto our robot. Our CAD team began fixing the geometry to incorporate this new prototype.

This prototype design uses two 4-inch Colson flywheels and four non-compliant flywheels. It stands at 26 inches with a speed of 6000 rotations per second with a 1:1 ratio at 100% max power with a Falcon. This would assist in obtaining faster cycle times, as they would be cut in half from our previous prototype, the rail shooter.

Our first test was at 50% power at 3000 rotations per second with a 1:1 ratio with a Falcon, 91.75 inches away from the Speaker, the closest our closest distance.

Our second test was at 70% power at 4200 rotations per second, 210.75 inches away from the Speaker, our farthest distance.

At 100% power and 6000 rotations per second, we effectively scored the Speaker 210.75 inches away from the Speaker.

At 100% power, were also able to obtain a full-field shot, extending past field dimensions to over 50 feet.

Above, we have the finalized CAD illustration of our robot for this season. Our robot is relatively small, at a height of 2 feet, 2 inches, to climb the chain efficiently while being able to score the Trap and be a defense-effective robot. Our robot’s size ensures that it will be steady on the field as our center of gravity won’t be directed to a specific area, and we will still be in the frame perimeters, not extending out more than a foot. The battery will be laid flat on the opposite side of our elevator, aiding with this concern. The shooter mechanism comprises the vertical double-flywheel style we settled for after prototyping. It consists of a feeder-roller above a tray that will feed the note to the shooter. The feeder-roller can hold the game piece while scoring for Amp or Trap without contacting with the intake, allowing for more flexibility.

Let us know if you have any questions. Have a wonderful day!

YYCDR · January 22, 2024, 3:44am

It appears you only have one “climber”. Are you concerned with the robot tilting?

Also what’s stopping the robot from tilting forward or backwards (depending where the cg is) when on the chain to score trap?

EHutson · January 22, 2024, 4:02am

Hi, So while we only have one climber in CAD right now the robot was designed to have one on each side. The second one just has not been added in CAD yet Regarding the robot tilting forward and backward we are placing the battery on the far front side of the robot to help. Our hope is to be able to ballast the robot so it is balanced. We are also able to change the location of our telescoping climbers to help balance the robot. If we do not have the weight budget for ballast we have discussed putting small wheels on the back of the elevator so we would be able to roll against the polycarb as we climb. Hope this helps!

David_IsMe · January 25, 2024, 3:40pm

For compression in both the horizontal and vertical rollers what did y’all find work best?

EHutson · January 26, 2024, 2:19am

So we are going with the horizontal rollers currently as we found that the horizontal rollers provide a more powerful and consistent shot. The compression that we are using is 1.5 between the outer diameter of the rollers, hope this helps!

reeti.praj4 · January 26, 2024, 4:34am

Much progress this third week!

We have started building our telescoping arms for our climber, which are still in progress. We’re aiming for compact arms in order to get us high enough to score the Trap successfully.

We finished building four v2 pinion swerve modules and two v1 modules tonight and plan on assembling two more v2 modules to add them onto the drivetrain by Saturday.

Our CAD team has been planning out the location of our electrical components in our CAD illustrations so our control subteam can reference these measurements for wiring to ensure that our robot’s weight is balanced. They have begun prepping electrical hardware by cutting wires to ideal lengths and crimping them for our robot.

We have finished cutting out our belly pan. We were originally going to cut out hexagonal shapes but resorted to triangular shapes instead, as we thought hexagonal shapes were too circular. The length between the edges of a hexagonal figure would be too short since we have to zip-tie our wires; a triangular shape was ultimately the best for form and functionality.

Our controls subteam has been working on updating our Scouting Website to this year’s game; this means updating the scouting forms, team data display, and other aspects. We added documentation to the code to maintain and update the site efficiently. Furthermore, we added a system to calculate a scouter’s accuracy, which was inspired by team 1678 Citrus Circuits. The link below will take you to our GitHub, giving you access to our code for our Scouting website.

Have a great day!

EHutson · January 26, 2024, 2:42pm

To add to this we have also finished assembling the intake pivot gearbox and have cut out most of the pieces for our intake. Our intake should be finished by tonight and we should have our drivebase wired by the end of Saturday!

We have also made a lot of progress with our shooter, most of the components have been cut and we are in the process of assembling it. Most of our major mechanisms will be done at the end of our Saturday meeting. We will have a lot of testing videos coming out soon!

reeti.praj4 · January 27, 2024, 11:11pm

Hello everyone!

We are making good progress tonight!

Today, we are focusing on drivetrain construction, which is still in progress. So far, it contains four V2 Swerve modules and the completed frame. We have almost finished wiring the drivebase. Initially, our Swerve modules were installed too tightly. We had to loosen the screws of the modules to allow for more leeway.

Our intake design with roller bars is underway. We encountered a few minor assembly issues, as some parts were assembled backward, holes were not big enough, and we lacked materials. Consequently, we had to redrill holes and rebuild components.

We rebuilt our umbilical cord many times. On our first version, we decided to dismantle it because we wanted to improve its design. We made a few mistakes while rebuilding the umbilical cord, so we had to dismantle it again. Our final product is shown in the picture below.

reeti.praj4 · January 28, 2024, 5:31am

We ended the night with much progress!

We began testing our final shooter design on our previous year’s robot and had successful results. In the video below, the shooter was tested 62.75 inches away from the Speaker at 60% power at 3600 rotations per second with a Kraken. The power of our roller bar shooters was not as high as that of our vertical shooter prototype, but we hope to increase its power once we add the feeder rollers.

Our robot has made significant progress, as by the night’s end, we had our shooter and elevator mounted onto our drivebase. For our elevator, we have rigging and the chain left to complete. Electrical components have been wired for the RoboRio, Ethernet, Radio, other CAN connections for the motors and sensors, and the general power for the robot.

More updates are coming soon. Have a nice day!

reeti.praj4 · February 10, 2024, 4:25pm

Hi everyone!

As competition comes closer and closer, our team has been putting more work into perfecting our robot. A lot was done this week- here are some updates:

Regarding our shooter, we scrapped our design of the vertical flywheel shooter as its height would have made the robot illegal when scoring into the Trap. If we were to implement this design, we would have to redesign our robot’s elevator geometry, which would not be practical to do at this phase with minimal time. We replaced the shooter with two-inch wheels and sushi rollers, which we tested on three hex rods. This is better than our original design with the vertical flywheel shooters because the new shooter makes the robot smaller and more compact, which is one of our top design priorities.

Sushi Roller Bar testing:

75% power (made into the speaker) approximately 60 inches away from the speaker.
100% power (did not make it into the speaker).

Furthermore, we had issues with our roller bar intake, which was built correctly but installed at the wrong angle. We realized that there was no specified pinch point, and the intake was minimally off-center on our drivetrain, causing us to have to redo the geometry of our intake.

Our concern with the intake design is that the shaft collars are larger, making it difficult to contain the specific gear that drives our Cancoder.

On top of this, we had concerns about other robots potentially hitting our rigging. To fix this issue, we are building a new plate to mount the beaker on, moving it forward to a more accessible area.

Moreover, we have encountered some issues with our elevator. The initial physical design of our elevator contained a hex rod extending over legal perimeters. Our bearing blocks were wide, causing us to stack them. This increased how much the shaft extended beyond the frame perimeters. These issues occurred as shaft components were not illustrated adequately in our CAD design. We also made further iterations, such as designing the elevator with thicker walls for better defense and a more robust overall structure instead of thinner walls.

We switched to a more rigid durometer flywheel. Then, we split rollers instead of keeping a single continuous roller to contact the game piece on either end. We will be swapping the Neo 550 for another Kraken, making us have three Krakens, two for the flywheels and one for the staging roller. Additionally, we are experimenting with pinching the piece between the staging roller and flywheel. This will be accomplished by reversing how we turn the flywheel to reach the Trap better.

Lastly, we decided to make our ratio higher. We are moving from a 36-12 to a 48-12 gear ratio because, with a 36-12, we can reach it from the maximum distance we will be shooting at. As we are running our Falcon at approximately 95-100% power to be more reliable as battery voltage drops throughout the match, it would be better to gear the ratio. Hence, it’s faster and uses less power on our motors. This allows for more stability for the shooter assembly.

reeti.praj4 · March 19, 2024, 9:51am

Hi everyone! It has been a few weeks, and we’ve been quite busy. Here is everything we’ve completed since we last posted

A few weeks back, we mock-tested our telescoping arm mechanism. We discovered that the note could not reach the Trap with our elevator. We originally planned on making the center stage of the elevator three inches taller to reach the trap, but we soon realized that the elevator originally extended exactly to its height of four feet; hence, if we were to increase the height of the stage, it would make our robot illegal as it would’ve gone beyond the height perimeters of 4 feet. Our top-level geometry had to be rebuilt and redesigned in CAD since we are now unable to score in the trap directly while climbing.

Our new shooter design consists of a horizontal bar shooter on a pivot, allowing for a flexible range of pivot angles to be determined quickly. It will enable the shooter’s pivot to align with the task. The design module consists of decreased compression on the stager rollers and added beam break to its functionality. This added soft limits to the pivot. Additionally, we increased the durometer of the flywheels to allow greater compression on the game piece so that it could be shot and scored efficiently.

As stated previously, we originally planned to design a robot that could score the trap directly after climbing the stage when the shooter would push open the trap. Due to the restriction of frame perimeters, this method was not achievable. With new iterations of the shooter, we are now able to successfully shoot into the trap without having to climb the chain by partially extending our elevator. We have had successful trials, although our performance on the trap has yet to be consistent. We are further prioritizing perfecting scoring for the Speaker and Amp, as well as our Autonomous routines.

2024 Trap Shooting Test

2024 Trap Testing #2

Several changes were made to our intake to resolve various issues that arose during testing. To maintain the lightweight and simplicity of the intake while protecting gears, gearbox integration was needed as it reversed the direction of the rollers into the space in between that supports the rollers. This iteration enabled us to make the design more durable.

In the event of damage, the intake’s pivot on the Spline XL shaft allows an effective and efficient swapping of the intakes’ ends quickly. The motor was set back to protect it from impacts during matches. Our design also added a beam break to detect when a note enters the intake. This enables us to flip the intake when a game piece is acquired.

Further changes were made to our intake, as our previous design caused a rough handoff between our intake and shooter mechanisms. We rebuilt the intake with the same overall design for a more efficient handoff that was shorter in length. This allowed for a softer grip on the note, allowing for higher scoring efficiency while simultaneously having the shooter pick up another game piece. Moreover, we designed the intake so the backrests on the bumpers can absorb force, making it more resilient to potential impacts.

|532.2368x217

Our telescoping climbers are from West Coast Products, which we found the most suitable for our design due to their minimal horizontal space usage and compactivity. The gearbox comprises the Kraken X60, which runs through a spool at a 20:25:1 gearbox ratio, and a dyneema rope, which can pull the telescoping arm up and down. They are attached to our frame rails using plates to ensure there is an optimal amount of strength to lift the robot successfully.

The video below shows our consistent and successful telescoping arm testing. At the telescoping arm’s complete compression, our robot is approximately 10.5 inches off the ground while climbing. In addition, our telescoping arms are at a fully extended length of approximately 14.3 inches while its compressed length is at 10 inches.11

Telescoping Climbers Testing
For our autonomous routines, we are using the program PathPlanner, an Application Programming Interface (API) that generates the trajectory for a robot to follow, ensuring smooth and accurate Autonomous movements. Additionally, we are using MotionMagic, a program used to generate trajectory and move to a target while setting out mechanisms to a certain position to make transitions between auto paths.

The video below presents our three-piece autonomous routine, which is currently in progress. As shown, our intake mechanism works simultaneously with our shooter to perform the handoff of the game piece. So far, it has shown consistent results.

3-Piece Autonomous Routine