Swerve Optimization

[aldaeron]

There are a lot more swerve threads this year than I remember in recent years and there is a lot of debate about the various designs. I would like to hear from the swerve veterans and the many designers: what are you trying to optimize with your design? I ask this as someone who loves design optimization, but knows very little about building a practical swerve module.

I see a lot of discussion around module weight and it appears under 8 pounds per module is “good”. Quickly comparing with a 6WD COTS WCD at Vex I get 22.01 pounds without fasteners (see below) compared with about 32 pounds for a swerve. This is actually quite amazing considering a swerve has 4 additional turning motors and their gearboxes!

I realize that swerve gives some pretty cool maneuvering capability for the ~10 extra pounds. And yes, I know that there is significant effort into learning to code and drive a swerve correctly on top of the added difficulty of custom machining 5+ modules during the build season.

As I said above – I want to know what the general focus is on with the many new swerve designs. What are you trying to optimize? Some ideas from the various posts (in no particular order):

• Weight of module
• Size of module
• CG of module
• Ability to rotate infinitely (i.e. not constrained by wires twisting)
• Gear Reduction available
• Two Speed modules
• More?

I am also interested to know at what point could swerve become COTS (yes I know about the Revolution Pro)? What would the price need to be? What would the weight/size/etc have to be for the non-diehard teams to get them?

-matto-

6WD COTS WCD

• 2X 2 CIM Ball Shifter (2.50 lb ea)
• 4X CIM Motors (2.80 lb ea)
• 6X Versa Blocks (0.22 lb ea)
• 6X Colson Performa 4"x1.5" 1/2" Hex bore (0.53 lb ea)
• 8X 9mm 18T pulley (0.03 lb ea)
• 4X 9mm 160T belt (0.07 lb ea)
• 1X 36” 1/2 Hex Stock (0.79 lb ea)

[jman4747]

I'm no expert but my priorities were (in order):

1. Cost
2. Reliability
3. Low Machining Complexity
4. Low Weight

[Tyler2517]

For last years it was.

1. Low machine time and complexity.
2. Efficiency (bearing over bushing ect getting the most per cim as possible)
3. Strength/rigidity
4. Cost
5. Weight

Also the foot print was played with. Most every thing was focused around how to make it as less complex as possible. One thing we did sacrifice was assembly... We did not think about that very much.

We also spend a lot of time in programming optimizing the steering algorithms along with some things to improve brown outs in the CAN bus.

[z_beeblebrox]

For the (tank) drivetrain my team is working on, our top priorities are basically:

-Is easy and fast to fabricate, assemble and maintain
-Is robust and rigid
-Is easily adaptable to different configurations
-Allows for mounting of electronics and pneumatics
-Has low center of gravity
-Has low weight
-Is compact

If we were ever to make a swerve, the priorities would likely be similar.

[asid61]

After actually attempting to create a swerve drive recently:

1. Machinability
2. Weight
3. Easy to assemble
4. Cost

I machine literally all the parts on our team, so I have a bias against hard-to-machine stuff now.
COTS swerve drives are prety much only available from 221, but that's fine. Their Revolution Swerve is very easy to incorporate into a decent swerve design.

To OP: For your 22lb estimate, you forgot to add the weight of the chassis itself. 0.1" thick tubing all around plus bellypan. Many of the recent swerve designs (particularly Aren's basic design) can be implemented without too much extra design work to make a whole chassis with 0.0625" thick tubing weigh around 34-35lbs. A 4-cim WCD might be a few pounds lighter for a very good one.
SWERVE 4 LYFE