Swerve drive cost

Hello and thanks for any replies,
My team and I are considering building test swerve module in the off-season. I was just wondering if anyone had any experience with how much a typical swerve module might cost. I currently have a design and it is somewhere around 400 dollars for one module. This number would get a little smaller if we built multiple modules.

I’m just going to drop the idea of “First Year Swerve Syndrome” into your vocabulary:

“Using the Engineering Design Process for Design of a Competition Robot, John V Neun”
Page 28, Section 1-a:

One common example of this is something called “FYSS” or “First-Year-Swerve-Syndrome.” This refers to what happens when a team tries to build a swerve drive for the first time. Most teams have difficulty building it the first time, and as a result they spend the entire season struggling to get it working effectively. Teams should be wary about building a swerve drive for the first time during a competition season lest they suffer from FYSS

Swerve on an individual level is going to cost ~350 depending on the complexity, and go down over time, looking at 1640’s Swerve Central page.

Swerve Central Breakdown

As your development ability improves, you may get it lower, but make no mistake. Swerve is expensive.

Swerve is hard. While you may be able to design and fabricate a swerve module for $400, you’re going to be spending a decent amount of time learning to program it. It might be worth while looking into a lower cost COTS solution for the hardware and focus more of your efforts on the software side.

I estimate the cost of our modules to be about $350-$400 a module. If you are willing to reuse components you can have very significant savings. Reusing encoders saves you $100 a module (assuming turning and drive vp encoders), motors around $40, and gears maybe $50. Pretty much the only non-reusable part of a swerve module is the machined parts. That being said there is a risk to not using new components on your competition bot that needs to be taken into account.

I did a breakdown of our swerve drive module cost here. It worked out to $290 for the COTS parts. We have since been looking into refining the design a bit. We realized that we could run VP Lite for the steering gearboxes which would save us $36.00 per module. Also, with the 9015s going away, we are planning to use Banebot RS-550s for the steering which will save us an additional $6.75. So, the adjusted cost is closer to $250 per module.

The rest of the module is waterjet cut out of aluminum plate. This is donated by a sponsor and I believe they typically use left over material from their own customer projects, so it is difficult to quantify the cost of that.

By the way, we have released both the CAD and the code for this design here on CD if you want to use them as a starting (or even ending) point for your off-season project. From a fabrication perspective, you need access to waterjet cutting capabilities and a lathe. Also, there is a bunch of details provided in this thread (including a video on how to cut the bearing grooves).

Our modules ended up costing us about $425 each for materials and all required COTS parts. Figuring out swerve in the off season is definitely the way to go. That’s what we did last summer to field our swerve last season.

$400 a module is safe, but remember that you’ll need to rebuild it in-season if you are successful. Also factor in that you’ll need 8 dedicated motor controllers for it, which is likely going to be around $400-800 all said and done.
As others have said, test it in the offseason first. If you don’t have 100% functionality by the time January rolls around, it’s not worth it.

Fabricate for $400 ok… but the design effort, both for the hardware, and then the software is going to be quite a daunting task that consume much of the team’s resources while you develop, test, re-develop, re-test and so on…

The cost of a swerve design project can be at least double what the end cost for a competition module would be. After worlds, we started to look at a redesign of our swerve. To use classic machining with metal for prototyping can be a long investment of time and expensive in materials. Additive manufacturing methods can greatly cut the time and cost of prototyping. We started off with a basic cad design. The parts were printed in cheap PLA filament. We actually designed several different versions. The modules are assembled, checked for fit and evaluated. Changes can be fast. Change the cad start the printer and be ready to reevaluate in a couple hour. While PLA is not suited for an actual module, 3d filaments have improved greatly in the past year. We are in the process of evaluating several carbon-PC, Carbon-nylon, carbon PETG and some newer PETG blends for toughness to stand up to a real competition.3d printing also has changed our design. Instead of thinking 2.5 d plated and minor true 3d, We have started thing true 3d integrated components. One still has to keep in mind what a 3d printer can actually print in 3d. So far we are very encouraged that a functional competition worthy module can be printed. That is our current goal for 2019. Of course the game will determine if we do it in 2019. Also, we don’t have a 15000$ printer. Not even a 3000$ printer. Our printer is a 550$ team designed and built. I will say it took allot of work to dial in each filament.