Almost have the steering modules done! Now I just need to put the ratchety good bits in. Top plate is solid steel (pulley is machined into it) with aluminum sides bolted on. Current weight hovering at about 3.5lbs. Any questions/comments/concerns?
Why steel over aluminum? What size wheel is that? And are the bevel gears in the little gap between the upper shaft and the wheel?
Looks interesting, although I’m not sure I would want to machine a pulley that tall with such tiny grooves.
Steel so it doesn’t bend as easy and shear teeth, 4" wheel, and yes. The pulley is only an inch tall…
Why are you machining the pulley on the “plate”? That’s probably the most expensive, hardest to manufacture option available to you. You’d have to start with a steel billet as big as the plate and tall as the pulley, then machine off 80% of it.
A pulley with a bolt circle and locating boss to mate with a bolt circle and locating hole on the plate would be easier and cheaper. Especially since then you could start with pulley rod stock and just machine the mating features.
I like where this is going, but how the heck do you plan to machine that pulley into the plate? You would be much better off purchasing the pulley and welding it on, or having it attach to the plate some other way.
I am aware this is a CAD and more of a thought experiment/practice than anything, but if you really want to practice and get better, you need to design things that are manufacturable. Unless you make a cast of that plate with the pulley machined into it, I don’t see a way you could manufacture this easily.
Other than that, is this a coaxial swerve/crab? How is the wheel actually powered (I believe I see holes off centered from the wheel, so I have no idea if there are bevel gears, or if it is not coaxial, etc).
The side plates seem to be significantly overbuilt, and that isn’t a big deal in this stage of the design, but it is something to keep in mind as you move further and further. In FRC, weight is a big deal, but in many other applications its really not a high priority.
How will this mount to a chassis allowing it to rotate? I am assiming there is a bearing or bushing somewhere that holds the side loads, and I would recommend you design that sooner rather than later, unless you already have a plan that is.
Other than that, keep it up! Iterate, improve, get feedback, and iterate more. I’m sure you will learn something either way!
Coaxial 6-wheel crab drive. The wheels are powered by bevel gears (the holes you see are the bearings in their holes). The mounting plate isn’t pictured, but it has another bearing for the drive shaft and bolts to the chassis via 6 1/4-20 bolts and held on to the steering module via shaft collar. Yeah the side plates are 1/2" alu. to protect the bearings/shaft/shaft collar from side impacts. Better overbuilt than broken 
Coaxial 6-wheel crab drive. The wheels are powered by bevel gears (the holes you see are the bearings in their holes). The mounting plate isn’t pictured, but it has another bearing for the drive shaft and bolts to the chassis via 6 1/4-20 bolts and held on to the steering module via shaft collar. Yeah the side plates are 1/2" alu. to protect the bearings/shaft/shaft collar from side impacts. Better overbuilt than broken
I am still confused about the bearings in the side plate. So one of the shafts goes through the center of the wheel, what is the upper one for then? If a shaft were to go through that it would also go through the wheel it seems. Mind explaining that a bit?
I’m glad you have already thought through the rest of the module, and I would like to see that once it is CADed up.
I do think that 1/2 in aluminum is significant overkill for this application. I get it, it makes it more bullet proof, and I am all for overbuilding than making repairs in the future. With that in mind there are many different ways you can beef this up without just using huge plates. The thing is the weak point in the system should be where these side plates attach to the plate with the pulley on it, not the plates themselves.
One idea might be to make a cross brace of some kind that goes between the plates, but ideally the shaft with wheels, snap rings, and the gear should be enough to keep the plates together.
May I ask why 6 wheels?
I understand wanting to make sure things never break, but you could easily take those side plates to 1/4" thick and it would still be plenty strong.
I’m also going to jump on the “have the pulley on top separate from the plate” bandwagon. The current configuration will just result in unnecessary cost and time.
Thanks for the suggestion! So one shaft goes through a pulley and the center of the wheel, while the other one has the other pulley and a bevel gear, but stops just short of the wheel.
Just going to repeat a sentiment that has popped up in your past threads before, because it bears repeating: I don’t see how having ratcheting wheels can possibly be construed as adding anything other than complexity.
I have three main things I want designing ratcheting swerve to achieve:
1.) CAD practise with complicated mechanisms
2.) Potential use in combat robotics/FRC
3.) Inspire innovation with a unique design
1 Like
6 wheel part definitely has to do with #2
All of these can be achieved without adding complexity for its own sake. Unique designs are inspiring when they provide novel or elegant functionality, not simply by virtue of their uniqueness.
Just my two cents.
Not exactly sure how thick the plate is but by pocketing it you are reducing the strength(resistance to bending) substantially for limited weight reduction. If you want similiar bend resistance move to a thicker 6061 plate( something around 1/4" with no pocketing).
With a pulley the torque transfer is spread over a number of grooves. A pulley or sprocket will normally have >1/3 in contact, in comparison spur gears have limited tooth contact
The belt is going to fail long before you shear teeth from your pulley, you can(and should) definitely use an aluminium pulley for this purpose( you could probably get away with a Markforged pulley if you increased the number of team or moved to a different type of timing belt).
If you are concerned about pulley damage consider using a larger pulley.
Yes! A simple, easy to machine and assemble swerve is going to inspire more innovation than a module that requires resources, experience, time and money that few teams have access to.
There is a certain elegance in simplicity. One of the things I say to the students constantly. Remember it when designing swerve.
So what I’m hearing is ditch the ratchet all together and go for stupid amounts of simplicity? There are already a few really good examples of simple swerve out there already…
What do you mean by ratchety swerve?
If you have a specific feature you want to add you’re better off adapting existing COTS or easily built design to add your feature, if possible.