FRC 9312 NERD Spark - 2024 Build Thread

Welcome to the NERD Spark 2024 Build Thread.

In case you missed it, we launched our CAD for our reconfigurable SWERVE chassis this summer. Thread is here:

We are still a 2 community team based both in Canton, MI and Albion, MI. We still planning on fielding two different programming teams, one that writes in LabVIEW and one that writes in JAVA. We’ve switched to OnShape for the most part but our mentor base is still more familiar with Solidworks so the transition is slow.

Here is our Website:

Github is here:

OnShape CAD is here:

Our sponsor included us in our town billboard so that was a treat for the team.

More to come but I’m already tire and I need a nap.

7 Likes

Week 1 has come and gone. NERD Spark has mimicked much of the rest of the OA teams’ progress. i.e. built shooter and intake prototypes, talked about overall strategy. There is very little that we can add to the conversation so I’m not going to post our boring prototypes videos.

I am struck by the luck of having come from FTC so recently, and having a similar game element in 2020. It was a much smaller orange foam ring but intaking and shooting required similar mechanisms. The team has reviewed the design in CAD and most likely will be using it for inspiration on our 2024 FRC robot.

For those who do not know the game, here is a video of our match. Keep in mind, this was the pandemic year and competition was virtually held at NERD HQ otherwise known as my basement.

For those interested here is CAD of that year’s robot.

CAD:

The shooter is powered by a single motor. A gear and pulley system connects the bottom and top wheels. The shooter wheels grabs one side of the ring giving it enough forward movement but also imparting just enough spin to keep the ring stable.

The design is similar to a lot of the really fast shooters shown already in OA. But most of these shooters have wheels all the way across the Note and thus does not spin the Note. I have yet to see anyone take out half the wheels to see how it affects spin and forward velocity. Our prototype for that is most likely a week away.

Design Strategy:

The team has finalized chassis size to a 24" wide by 28" long robot. Using our reconfigurable OnShape CAD model, we generated the DXF’s and sent files to our sponsor’s laser last Wednesday. Parts arrived on this Monday. We sanded and painted everything black. We specifically use this:

"Rust-Oleum 7220830 Textured Spray Paint, 12 oz, Black

which can be found locally at home depot. The spray paint covers well and dries to the touch in about 10 minutes.

We’ll be wiring all week and the LabVIEW team will hopefully have the chassis shaken out by the weekend.

24X28 size was chosen based on this design strategy

It is a combined intake shooter which should only take about 60 % of the real estate of the chassis. That will allow us to intake from the ground and shoot into the Speaker. To handle the Amp and possibly the Trap, we’ll implement a super lightweight 2 DOF arm with an additional wrist joint/gripper. Gripper will grab the Note directly from the shooter opening. If we have time, we’ll implement a shooter to arm pass off that actually has the shooter shooting the note 1 foot across the robot into the gripper of the arm which will then dunk it into the Amp or Trap. Not saying this is the smartest thing to do, but it would be pretty cool to pull off.

Turd Swerve
Lastly, We wanted to share our Swerve Training and Development platform. The idea is to develop a super cheap fully 3D printed Swerve Pod but use FRC legal motors and electronics so that the software could be transferred fairly seamlessly to an actual FRC robot. One use case is to allow pre-swerve teams the ability to try out swerve without buying expensive swerve pods. The RIO, NEO’s etc, could all be re-used for eventual competition robots if they choose not to do swerve so the added cost would be under $100 of general bearings and parts. Another use case would be to allow SWERVE-enabled teams to use this platform to teach younger members of the programming team. It might be too expensive to provide every member of the programming team with an FRC sized test bot, but this platform would be cost effective enough to have a few among the programming group to test sensors, auton, and other algorithms.

The design was meant to be ultra cheap so definitely not meant for any competition use. It’s not quite up to the name NERD Swerve, so we decided to call it TURD Swerve instead.

Here is a video of it running around:

OnShape CAD:

Github:

As always we welcome any questions or comments.

That’s all we have for now. Happy week 2.

7 Likes

FYI, Mag Encoders on the swerve pods are using Thriftybots:

Integrated fairly seamlessly and performance has been rock solid.

1 Like

Shooter Prototype:

Using 4 Thritybot Solid Urethane Wheels:

Powered by 2 NEO Vortex in open loop control.

1/2" compression.

What’s slightly novel is that the rollers are only on one side of the note. It creates a fairly smooth spin but at the cost of overall forward speed.

Shooting from the back of the wing, this video proved to be consistent during this run but in subsequent runs was less so.

We’re implementing closed loop velocity PID control to improve consistency. A lot of the consistency issue is from variations in feeding the note as many have pointed out.

27 Likes

This video is fantastic and thank you for sharing. I never thought to do a top-bottom launcher on just one side.

1 Like

Cool! Looks super promising

2 Vortex open loop at what gearing? 1:1?
And open loop at 100%/12V?

1:1 gear, direct drive. And yes, open loop, percent vbus, 100% output.

1 Like

Can you share the angle from the shooter? And the distance from the floor

It was about 35 degrees. And it was only 5 or 6 inches from the floor.

1 Like

Wall Rider

59 Likes

This is one of my favorite build blogs now.

10 Likes

I wouldn’t want to build my robot around this due to note degradation and it’s not much (if any) of an advantage on the surface, but this is freaking brilliant all the same.

5 Likes

Love the work and wild ideas! Keep it up!

5 Likes

I love this more than I should, but you should be aware that there’s a piece of aluminum angle at the top of that black wall that might slow your roll.

3 Likes

this looks amazing! is it as consistent as a traditional shooter, or are there any extra steps needed for things like alignment?

Do you have a photo of the shooter or a link to the CAD? THanks

Looks to me like it is their single sided top/bottom roller shooter just rotated 90 degrees so there is forward spin.

1 Like

We are not endorsing the shooter design. It was just funny and interesting. There are a lot of reasons this design is sub-optimal.

  1. Easy to block.
  2. Needs a very specific bounce and thus probably not going to be consistent or at least will need tons of tuning.
  3. We haven’t tested an off angle shot but my guess is that it would not work well.
  4. Intake/indexer now needs to rotate the Note 90 degrees making it more complicated.
  5. Shooter takes up vertical space, which is counter to the design goals of staying low.

We don’t have models of shooter yet. It was pretty hacked together.

We essentially mounted NEO Vortex’s on one side and then 3D printed these pillow block bearings on the other side:

We used Thirftybots’ 4" solid urethane rollers and moved them along the hex shaft to study how well it developed spin and how well it created forward movement.

7 Likes

We’ve been radio silent for a number of weeks. We’ve been pulling some really late nights and I haven’t had the stamina to stay up and post.

CAD finished about a week ago. Link above. And we’ve spent the past 2 weekends building, testing and debugging and programming.

We hope to have a completed bot by the end of the week which gives programming the weekend to complete the autons and for drive team to have a couple days to do some hard drive practice before our week 1 at Jackson.

A quick video of the robot with shooter/intake fully programmed.

We need to fix a bug in programming to make sure driving is correct. And we need to implement programming on Arms and climb.

Questions and comments welcomed as always.

38 Likes

Shooter intake combo on a 4-Bar!

Looking good.

2 Likes