custom wheels

[ablatner]

118 has cool custom wheels. They 3D print the hubs, but then mold the tread themselves. They originally 3D printed the tread, then made the appropriate mold based off that.

[MasonMM]

Originally Posted by ablatner

118 has cool custom wheels. They 3D print the hubs, but then mold the tread themselves. They originally 3D printed the tread, then made the appropriate mold based off that.

Almost. We actually mold our two piece hubs out of polyurethane plastic. The hub is designed to clamp our custom treads, which are molded from a concoction of polyurethane rubbers. A positive of each element is grown on the 3D printer and we make the molds by pouring silicon over the grown part. We can then make unlimited silicon molds from a single grown part.

[ToddF]

Originally Posted by 1683cadder

I noticed that a lot of teams are using custom wheels this year. Is there any advantages? What about costs? And what is needed when designing a wheel?

Last summer our team did some drive train development and on our robot this year we used a west coast drive with 4" Colson wheels:
http://www.robotmarketplace.com/products/BPDWC09.html

These were mounted on aluminum hex hubs:
http://wcproducts.net/wcp-00053/

We were absolutely thrilled with them. They have some big advantages:
- Being COTS components, they can be pre-purchased. Other than a quick sanity check to be sure you aren't making a huge mistake, the cost in build season time is essentially zero.
- They are very durable, and have very high traction.

Now, the costs:
- Each wheel costs a total of $20.99, so a set of six costs $125.94
- Each wheel weighs a total of .634 lb, so a set of six weighs 3.81 lb

With a baseline established, you have a basis for improvement.

Our team purchased a Makerbot Replicator just before the build season started, and used it to make things like hex shaft spacers, gearbox covers, and camera brackets. So far, strictly non-structural parts. We have been inspired by other teams to begin experimenting with making structural parts. We have started by looking at wheels, mostly because making your own wheels is cool, and that's what the off season is about, having fun and playing with cool things. We think it would be cool to have our team logo, or at least lettering, on the face of our wheels. So, we are playing a bit.

We started with this wheel.

Pretty cool looking, but nowhere to put a logo or lettering. Also, it could easily be just CNC machined. It doesn't take advantage of the nifty abilities of 3D printers.

Next, we did this one:

Can't machine this with a CNC machine. The spoke size and rim dimensions can be modified to make the wheel stronger, if testing shows that more strength is needed. But, the spokes are vulnerable and there still isn't anywhere to add a logo.

This is the latest wheel (printed last night):

Internally, this one has the same spokes as the previous wheel. On the inside and outside, there is a fairing surface that protects the spokes, adds some stiffness (and weight), and provides a surface on which to put a logo.

So let's do some comparisons to our baseline.
- Each wheel takes somewhere around 6 to 7 hours to print. This means that if they printed around the clock, a set of 6 wheels could be ready to go on the bot on Monday, after kickoff.
- Traction and durability are as yet untested, but based on the experiences of others, we should be able to make the hubs durable enough, provided that we do proper testing during the summer. Traction and durability of the treads will be on par with all the other wheels out there that require strip treads to be added.
- Cost: This can be looked at one of two ways. You can either include the cost of the machine and the plastic, or not. Because we already have purchased the machine and plastic, and the price of the plastic on the BOM is very low, I consider these parts to be free to the team. A 10 foot piece of 1" wide tread is going for $24 at AM right now, so we'll take $25 as the cost for all six wheels.
- Weight: The last wheel hub shown weighs .21 lb. Tread weighs .11 lb per foot (which is about the length needed for a 4" wheel), so take the weight of one wheel as .35 lb, and 2.1 lb for a set of 6.

So, those are the potential advantages and costs, from our teams perspective. One benefit not listed is that the team gets the experience of designing, testing, and building something cool. That's not worth nothing, even if it doesn't go on the robot.

What is needed to do the design?
- A 3D CAD package capable of generating stl files. Also helpful is some FEA capability, to help you make design decisions about geometry tradeoffs. Experienced design engineers can get very close to good designs by eyeballing things and using their judgement, but high school students don't have the experience yet. This is where working with your mentors will save you time.
- The initiative and enthusiasm to pursue the project.
- Access to a 3D printer, to build prototypes.
- Time and equipment to test the prototype wheels. We plan to torque test our wheels to failure, to see what their weak points are. We'll also torture test them on a robot to see if they will hold up to real world wear and tear. You want to break them to find their weak points now, not during competition.
- The willingness to take the results of all your hard work and publish it for the rest of the FIRST community to benefit from. Lawyering the rules aside, for you to fabricate something from a design you have developed before kickoff day, you have to publish the design for anyone else to use.

[Aren_Hill]

Originally Posted by sentientfungus

On another note, one of the benefits can be an integrated pulley, such as 3928 had on this year's robot. I can't tell you how nice that looked for a team that was struggling to use belts for the first time on our drive train this year.

Can't have a pulley come loose if its part of the wheel

We had zero issues with our drive modules and the belt running to the traction wheel, We managed to get them setup for a single CNC operation, and then slice off on the lathe. Very quick operations, worthwhile overall.

[EricH]

Originally Posted by Tmaxxrox97

Actually a RepRap printer that would be able to make a pretty decent wheel is around $500- $1000 depending on where you buy the parts and on which model you buy. That is also an un-assembled version (I'm assuming that a robotics team would be able to put it together).

Won't argue on cost, but I would suggest that the ability to put it together might run into some issues. I've built a couple RepRaps (actually, somewhat modified versions) and if I hadn't had a fully-functional one to reference, I don't think I could have done it. Mechanically, probably. Electrically, bring your own reference documentation.

Then I had the pleasure of showing a bunch of other teams how to build these printers. It took all day to get the mechanical part down; the electrical was being started.

I actually went in and rewrote the entire assembly instructions--CAD included--to make them easier to understand. (Couldn't do anything about the electrical, though.)


Now, that wheel comparison shown is pretty solid. Saving a pound and a hundred bucks, and possibly some time, could be huge in this competition. I do encourage that testing you're planning to do--break it, rebuild it, break it again, work all the bugs out, then build one without the fairing on one side for the judges to see.

[DjParaNoize-]

This year we designed some Custom 4" wheels that were made up of:
Two Custom Designed Rims That Were then Machined By Our Sponsor Chevron.
The Overall Idea was to be able to switch out The Fabric Quickly for when they would wear out. But it ended up that the thread we used didn't wear off that much!
We got a Roll Of Yellow Polyurethane and Got it Cut To Fit Our Rims which was a great thing!

Lots of People Would Come over to the pit and ask about our Wheel and look at it, so I decided to share some of the information here.

Images of the Rims: