Molding Custom Wheels?

[Cory]

Moldings represent a very large investment, both monetary and in time.

You will need to machine a mold for your wheel. This is not simple, and will cost a LOT of money for you to have a shop make (Likely more time than if you just CNC machined however many wheels you want to make).

You'll then need someone to injection mold your wheels. This will also be quite expensive.

Molding is really not something you do for onesies and twosies. If you're getting something injection molded, you're making thousands of them. Any less, and you cannot justify the cost involved.

[MrForbes]

Originally Posted by Cory

You will need to machine a mold for your wheel. This is not simple, and will cost a LOT of money for you to have a shop make (Likely more time than if you just CNC machined however many wheels you want to make).

I haven't looked into it, but I understand that you can make molds with the help of a 3d printer....might cost quite a bit less than CNC machining?

Of course I would just redesign the wheel so it's easy to make.....

[Akash Rastogi]

Thanks guys, I think I'm going to make another revision of the wheel that can be CNC'd . I might try to get one done on a rapid prototyper though. That'd be a cool thing to try out (an alum at MIT can do it for me).

Any and all other suggestions are welcome.

What problems would a CNC face while trying to machine this with the curvature of the faces? That's the only thing I don't know about since I personally haven't done anything on a CNC yet.

[Cory]

Originally Posted by Akash Rastogi

What problems would a CNC face while trying to machine this with the curvature of the faces? That's the only thing I don't know about since I personally haven't done anything on a CNC yet.

It would take a really long time. fillets in inside corners are no problem, but fillets of faces are difficult. You either need to be able to use a corner rounding end mill, or you have to 3d contour the surface to get the desired radius. In general avoid corner rounds on any machined parts whenever you can.

[Brandon Holley]

now now now...lets not jump ahead of ourselves here.

Akash,
By molding....what do you mean? Cory explained spot on the difficulty with injection molding. Money, time, actual injection...feasibility wise I would say close to impossible for a FIRST team, on any budget.

However, I've made wheels before by molding them myself. Silicone and urethane are 2 wonderful materials you can use. Making a silicone mold is fairly easy...getting them to fill properly is a lot more difficult, but not out of the realm of possibility for anyone with some kind of mechanical inclination.

I'd say for almost any wheel you make, your going to want to use a split mold. If you can gain access to a 3D printer of some sort I can tell you a VERY easy way to make a split mold (This is what I do all the time at work).

Once the mold is done, you should be able to get 10 or more wheels out of it as long as you take care of it properly.

You could go as far as to mold the hub, and then mold the rubber around the hub in a separate mold (yeah that's right, no nitrile). The nice thing about this is you can dial the durometer to exactly what you want.

If you have any more questions about this feel free to ask me and I'll gladly help you out.


One final note...keep in mind, if your making a wheel out of some sort of plastic, it is going to be lighter than an aluminum wheel (if you make the same exact shape), however it may not be as strong. Keep that in mind...you may need to balance some of your weight savings with material to your weight savings of mechanical design.

-Brando

[JesseK]

Team 357 (The Jesters) has some experience molding smaller parts. They've made their own custom mecanum rollers for years now. If you have a copy of the FIRST: Behind the Design books, they're in 2007 (I think) and describe how they made quick molds for things like their battery holder, mecanum rollers, and even a 80/20 UHMW plastic linear slide.

[EricH]

Originally Posted by JesseK

Team 357 (The Jesters) has some experience molding smaller parts. They've made their own custom mecanum rollers for years now. If you have a copy of the FIRST: Behind the Design books, they're in 2007 (I think) and describe how they made quick molds for things like their battery holder, mecanum rollers, and even a 80/20 UHMW plastic linear slide.

They make the whole wheel... And they're one of the few teams to be in both Behind the Design books. 2006 details their wheels; 2007 goes after everything else. As I recall, they also went for a patent on the linear slide (which isn't UHMW; it's urethane, polyester, and a molybdenum compound molded into shape).

[JesseK]

Originally Posted by EricH

They make the whole wheel... And they're one of the few teams to be in both Behind the Design books. 2006 details their wheels; 2007 goes after everything else. As I recall, they also went for a patent on the linear slide (which isn't UHMW; it's urethane, polyester, and a molybdenum compound molded into shape).

I thought the patent was for their battery holder.

Meh, it's in the book which is at home.

[EricH]

Originally Posted by JesseK

I thought the patent was for their battery holder.

Meh, it's in the book which is at home.

Double whammy. They were going for both, I think. *goes to the other room for family copy of 2007's book* Yep, that's 2 patents, one for the battery system, one for the linear bearings.

[JesseK]

I spoke with some guys at a conference yesterday and had the pleasure of discussing robotics design with a guy from American Plastics Manufacturing based out of southern IN. They specialize in injection molding. There are a few rules of thumb he described and this particular wheel appears to break two of them: Keep a UNIFORM wall thickness, and have NO sharp corners.

Injection molded plastic has a flow direction, and all polymer compounds shrink 0.5%-3.0% depending on the compound. Most likely you'd go with a crystalline-type plastic material for the wheels, which also has unpredictable shrinkage. This shrinkage causes a variable wall thickness injection to shrink at different rates, thereby warping whatever item you're trying to make. When the variance is too great, many times the thicker area of the part will crack since the plastic is still trying to shrink after cooling down. In addition, a sharp corner will not have plastic flow smoothly through it, which will cause a non-uniform shrinkage at the corner when it cools. This severely weakens the corner.

He had some equations up, and depending on the strength of the plastic material you choose for the wheel I'd estimate that you'd want a solid 1/4" thickness all the way around, with (and this is critical) no lower than a 1/2" radius on the fillets. If you use a non-crystalline plastic, then you can have a slight variance for your thicknesses, but not much. AM's wheels are ingeniusly designed and have the right material composition. Though as a warning, since polycarb is non-crystalline, I now think that teams should take the weight of the bot off the wheels while it's in shipment, since the shipments can go through various extremes of heat which will cause the polycarb to creep.

Hope this helps if you still decide to do molding. -- edit -- As a followup, if you do want to do the molding, follow the link above to the conference and get the contact info of the APM guy. He was glad to give me a couple of pointers, and probably wouldn't mind spending a few minutes looking at your design.

I'm the one who voted other.

Do both. Make your design so it can be CNC'd, then rapid prototype some lovely little molds if you feel like doing mold work. If not, you now have a design that is either easily interchanged, or is dual function.

[FourPenguins]

So my post just got eaten. I will try again.
Pros of molding (using Urethan and silicone as Brandon suggested):
-Reproduceability
-Can be done in a garage once the molds/dies are made
-Doesn't tie up machining resources
-Is a fun job for freshmen
Cons:
-Lacks strength of aluminum wheels
-Less material flexibility
-Improperly mixed resin will have very poor structural properties (worst case is putty)

A quick overview of the process: two parts are mixed, a resin and a hardener (as in an epoxy). This is poured into a premade mold and allowed to harden. You then break it out of the mold and debur with a sander/file. Brandon's idea of molding silicone right onto the wheel as tread sounds pretty awesome. For mold design and process questions, hit up Brandon, me, or Mike. Mike and I worked with this stuff back in 2005 and apparently Brandon now gets paid for this
For material ideas see SmoothOn's website (The same company that makes the urethane foams ubiquitous in MORT robots.)

[rajikurbaj357]

Originally Posted by EricH

Double whammy. They were going for both, I think. *goes to the other room for family copy of 2007's book* Yep, that's 2 patents, one for the battery system, one for the linear bearings.

I know I'm a little late, but we (357) have a patent on both, the battery base patent was just approved a few weeks ago.