Here is a 2-speed swerve concept. I wanted to see how low I could get the weight - CAD says about 7.4 lbs (without sensors.) It would have a bit more machining than I’d be comfortable taking on in a build season.
The outer diameter on your tread and the gear teeth on your gear look very close. Not sure what the distance is pictures can be deceiving, but running on gear teeth would be a bad thing.
Also are you using a worm gear for turning the module, I think I can see a little bit of one through the reflection. Could you post a side view so we can see the turning motor? I like the tubular module.
Yes - the gear is risky. I used a gear that was the same size as the rib of the wheel on the other side. For this study I was trying to see how far I could push everything. It’s fun (an engineer’s fun) to take these risks when it doesn’t really matter. In the middle of a build season you can’t get away with stuff like this when you don’t know for sure that it will work.
I used an RS-395 motor with a 40:1 worm and a 4:1 gear set. I don’t have any experience with swerve turning so this was a bit of a guess. You have sharp eyes to pick this up from what you could see in the picture.
My concern would be that the steering gear is too far from the support for the module. Under side loads the gear could potentially be caused to disengage or bind if there is any play in the module. By moving the gear up until it is as close to the bearing as possible, the gear spacing will not be as affected by side loads, even if wear increases the amount of play in the module. Alternatively, you could add more support below the steering gear like on the wild swerve module. This has the added benefit of a greater mechanical advantage to resist any side loads.
I have a Kaydon bearing at the top of the wheel module and a nylon ring supporting it just above the steering gear. I would have liked a second Kaydon bearing right at the steering gear but the size didn’t package very well. Trade-offs for later if we did pursue it.
Where are your sensors?
I would say go for it in the design. I personally think that your turning gear placement should be fine. From the picture it seems like it could be weak but this is a module, and not an entire frame so I can’t say that when implemented in a frame that it wouldn’t work. I really like the worm gear idea very interesting. I try to avoid using a banebot on a drivetrain because I really like their functionality for other areas of the bot. But that is all preference really.
I know what you are saying, but when the thing is built you could have a lot of people frustrated with the design, say they want it changed, and you are forced to go back in and fix something. I would look into timing belts (a lot of people don’t like them) but in a swerve drive they can be very effective. http://sdp-si.com/ has a really cool belt length calculator to make designing easier. Also are you using CREO? You can also get into equations and make the design change according to what reduction you prefer on the belts.
Just food for thought looks very sharp.
There are none. If we did pursue it, I would get the electronics / software people to pick the sensors and then work with the students to find placements for them.
This concept design is intended to bolt easily to a C-channel frame. If we did a competition version we’d have to think about how it would really look.
Yes, I we are using Creo.
Curious for a more detailed picture of the module itself (just the rotating part) and the bearing mount.
I’m always up for seeing another way to make a swerve module