Swerve drivetrains of increasing impracticality

Finding myself with an abundance of time to kill over quarantine, I decided to learn CAD. During this impressionable time, I saw hmccree’s post regarding impractical WCD drivetrains and decided to try my own hand at it. However, wanting to put some of my own flair into it, all of these drivetrains utilize the WCP swerve modules. Secondly, as you will see, I skipped by making any ‘practical’ drivetrains and straight to the impractical.

Swerb V1
My first drivetrain answers the question that I am sure everyone has asked: “what would cambered swerve look like?” Not much to say about this guy that the renders don’t. It features -15 degrees of camber. As you can see the chassis does not look very rigid. This could be remedied with a bellypan and/or adding further 2x1s horizontally across the non cambered portion of the chassis.

Swerb V2
I then realized that a 4 falcon drivetrain would likely not have enough pushing power, so I simply added 2 more swere modules. This chassis fixes most of the rigidity issues in v1 by adding further supports for the extra modules.



Swerb V3
My final iteration, this drivetrain features some minimal improvements regarding lightening and gusseting. Camber angle was also increased to -25 degrees. It was around this point that I began to consider the practicality of these drivetrains. I then realized the turning may become difficult with the 6 wheels. As such Swerb V3 features a 0.25" drop center for ease of turning.

Apologies for the weird renders, I am still fairly new to CAD.

Feel free to ask me any questions or give feedback.

Drive link to all renders and step files: https://drive.google.com/drive/folders/19p2jnJydWGtazsyo0v0G9U1-QnL7AUNu?usp=sharing

16 Likes

4/10

Not enough throttle motors

11 Likes

I get the impracticality part, but besides the obvious, “Why?” what would be the benefit of a camber on a swerve module? It will make it more difficult to hold position, to rotate, and you kind of roll over the side of the wheel as the module rotates.

Camber on a WCD kind of makes sense, mostly because the wheels are in a fixed orientation relative to the floor, whereas a swerve wheel completely changes orientation when on a camber.

1 Like

While this is really cool, I have always preferred fully symmetrical swerve drives. I think the next step in innovation is to have 8 swerve modules, making full use of the available PDP slots. After all, the drivetrain is the most important mechanism on the robot!

9 Likes

To do list:

  1. Need to design a round frame to go with my round bumbers (highly impractical since round bumpers can only be made legally through highly impractical manufacturing techniques that also render them structurally inadequate for their job)
  2. Incorporate an odd number of swerve modules (5 or 7) preferably with an uneven spacing around the frame just to give the programmers nightmares.
  3. 6 swerve drivebase with a drop center
  4. Omni-swerve

Edit:
5. H-drive omni-swerve (imagine the raw juking power)
6. Mechanum swerve (would that be an automatic DNP?)
7. 8 swerve drivebase (thanks @AveryP)
8. 10 swerve drivebase (really impractical since there are not enough PDP slots)

6 Likes

I think you’ll like this

2 Likes

I made this because of the impracticality of it all. Swerve and camber together is a bad which will likely result in burning out multiple rotation motors.

1 Like

Idea: combine these two and have 5 outer swerves with uneven spacing and a 6th swerve on the inside with a drop (but of course not in the exact center)

6 Likes

Maybe use a spherical wheel profile to maintain constant wheel contact.

1 Like

But then you lose the increased traction you can gain with 11-sided wheels!

3 Likes

What about a mecanum with only one swerve in the middle?

1 Like

Please drive a pair of swerves through a differential which of course said diff gearbox should be powered by 2 falcons.

2 Likes

angle swerve


Credit to @MisterRocketMan (Lazro) on the FRC Discord

10 Likes

Drop-center H-Drive kiwi octocanum swerve

1 Like

Alright, this was something I was working on last year that I think fits here. I decided it totally was not worth designing, but I hope you get the general idea and a good laugh.

This uses a 60 deg bevel set instead of a 90 deg to create the angle. There really is no benefit to this over a standard drive besides maybe a slight efficiency improvement because of the decreased bevel angle. It would look really cool though.

12 Likes

Why use the wcp modules when you could slap on a half dozen of these:

2 Likes

I think to increase performance, you need to camber the swerve on both sides so it sit’s diagonally no matter the orientation of the robot.

Team 190 actually built a canted swerve in 1999 and 2000:

4 Likes

Just to piggyback off of the “somebody actually did this”, 1625 made a 6 wheel swerve in 2010, complete with a drop center (and if nobody did it after them, the “100% of 6 wheel swerves make it to Einstein’s” still holds true).

It helped with reorientation of the robot, which was a greater issue back then.

7 Likes

I forgot to add my done list:

  1. Swank
3 Likes