Apple Pi Offseason Swerve CAD and Labview

Apple Pi, Team 2067 is pleased to post our offseason swerve mechanical design and Labview programming files.

This past summer we designed and built 2 different configurations of modules, both featuring 3d printing. After that in the Fall we designed and built another which is presented here.

The swerve module features:

3D printed pivot pieces (4) + 2 simple 2D plates, the rest is all COTs.

The Lamprey encoder (221 systems) is incorporated to directly measure the pivot rotation

3.0” ID, 3.5” OD main bearing from TheThriftyBot (aka FRC Costco)

Falcon Drive Motor (prototype made with the Neo because ….)

RS550 steering motor with a 69:1 ratio (BAG or 775pro also can work with minor tweak to plate)

3 in. Colsen wheel (or 3” 3d printed treaded wheel)

AndyMark 2:1 bevel gears

Currently geared for 14.45 or 12.85 fps (theoretical with the Falcon)

4lbs. 1oz. total weight, (with NEO drive and not including the 2 motor controllers, but everything else)

Currently 3.25” from frame edge, and 8.1” tall with Falcon (7.2” with NEO). Though we really like the Mean Machine closer to edge configuration – which could get this to 2.25” from edge.

Labview Programming:

Offseason swerve code, can support either Sparkmax or Talon SRX for drive, Talon SRX for steering (motion magic), Pigeon IMU, Lamprey encoder. For anyone desiring a more generic Labview swerve (all on RoboRio, and an analog encoder) please look at our 2015 swerve code, searchable on CD.

We also are including our Labview training package, accompanied with a Labview project. This was used for training of our mostly new programming team this Fall, and is used to augment the Secret Book of Labview, which is great for introducing Labview, but we add in some examples for incorporating the Talon SRX features. This is a work in progress, more features will be added next year, along with better commenting in the project files.

Apple Pi 2019 Offseason Swerve Assembly.zip (10.1 MB)

2019 Training Robot Project.zip (11.6 MB)
2019 Offseason Swerve Update.zip (304.0 KB)

8 Likes

What did you print the printed parts out of? This looks really cool, I’m glad to see the lamprey getting some use!

ABS for the prototype, but would likely move to Onyx or similar if use in season. The ABS held up in our testing but I wouldn’t want to have the season riding on it.

Are you considering replacing the rs550 with a neo550? The module could be made to be shorter and it also slightly reduces wiring complexity as you only have one type of motor controller instead of two. Depending on your steering backlash, you might even be able to use the neo550’s encoder for steering rather than your current Lamprey encoder.

The reason for the external encoder is because it spins 1:1 with the azimuth. With just the built in Neo550 encoder, during startup, would not provide the exact position of the azimuth. For example if steering was done on a 69:1 ratio, the Neo550’s encoder reading could correspond to 69 different azimuth positions, all 360/69, roughly 6 degrees apart.

I’d like to note that it isn’t impossible to run a setup without an external encoder, the module would need to have a way to localize the initial reading i.e. magnet + hall effect sensor.

Could you swap out the printed parts between competition now that bag day is gone? Also, how many hours did you have on your swerve?

We’ll see, when the neo 550 ever comes out then can evaluate - however we will probably prefer to stick with the CTRE controllers as our steering code uses its motion magic feature.

When we have gone with swerve we have always had two sets of modules, however we have never had to swap one out. We would not go in planning to swap out, but would be prepared to. Swapping out a module can be done in about 5-10 minutes.
We don’t have that many hours on this configuration - but have about 10 hours on a similar 3d printed versions - including running on rough asphalt, which shakes it up pretty well. Again this was with the ABS material, and we would likely upgrade that, if we go this route. Last year we ran with a nylon 3D printed main pulley/bushing (instead of bearing) and it held up fine, including running off of the Hab.

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.