Hey, I’ve been trying to design a swerve drive for my team but am coming across some design challenges. You see our team likes to beat up our bot, so we tend to use a robust serpintine belt driven west coast system. This is one of my main problems because I don’t think we can give up that robustness. So I’ve been researching mounting methods for the modules that would be sturdy, the first thing that comes to mind is 2910’s method where they cnc a pulley and slip it inside a silverthin bearing, next is the revolution swerve module which I honestly need some one to explain to me but alot of people have been saying it’s one of the most robust designs. Finally is one that I have seen made by @asid61 in the gbx-136 which has a thrust bearing built into the gear. With this design It looks robust but I’m concerned that the small bearings would sheer their bolts with a large impact. Please provide suggestions of any other ways I could mount the twisting portion of the module. Thanks!
First of all, realizing the importance of robustness this early is a great start, many teams realize it much too late. I can attest to the rev modules being robust, we ran the same ones for 3 seasons (including 2017 when our robot dropped directly on them repeatedly).
To understand how the revolution modules work I’d recommend pulling up their cad and picking it apart, that’s how I learned. Also on the presentation linked below on slide 6 there is a cross section of 2767s modules which have a similar approach.
if you want a written explanation of how they work here is my attempt (its way easier to explain in CAD): essentially you have 2 flanged bushings. One of them is static and attached to some sort of plate, the other bushing is either inside or outside of the static one and rotates around it. Attached to that bushing are both the pulley to rotate it and the drive wheel assembly. The drive shaft goes through the middle of the inner bushing with a bevel gear on the end to power the drive wheel.
If you have a design you’d like feedback on feel free to pm me.
Our team is also in the same boat; we’ve had an aspiration to be able to do a swerve drive but the times we’ve tried it never comes to fruition. I’ve looked into the resources available and it looks like there is a module that Andymark sells for swerve. Is this a viable option or is it more beneficial to just design one ourselves?
Designing your own modules is always the best option because it is up to you to integrate them into the rest of your chassis however you like. However, off-the-shelf swerve modules are definitely not off the table. My team (3082) used premade modules this past offseason and are continuing to use the same style this season. We would like to design our own modules in the future since the premade modules are not an ideal shape for us, but we did not have the time to revise and manufacture our own for this season. So I would say design if you can, but premade modules are certainly a viable option if you plan the rest of your robot around them.
The Andymark modules are limiting in their size/weight factor and lack of a drive encoder.
GBX-136 had a massive weakpoint in those small bearings, one that I didn’t catch when I made it at first. The bearing setup form GBX-162 is worth looking into instead: pic: GBX-162, Realistic Swerve
That’s the same style that 1323 uses on their modules.
WCP also has a COTS swerve module, although it’s not in stock yet.
@asid61ok I see what you mean but where did you source that large gear? I’m wondering because our team (2811) uses exclusively 20dp gears from vex and andymark
As a mentor for 1640, I would suggest you look at our “ swerve central “ site. It might help in your thoughts and implementing your swerve drive. We have cad and programming for swerves. Team 1640 swerve central.wiki. We have 7+ years of info for you to pick from, including our cvt swerve
I’ll check that out thanks for the advice
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.