pic: Why aren't omni wheels like this used?

[Adrian Clark]

Originally Posted by Ether

Having never had the opportunity to see a DuraOmni up close and personal, I'm interested in understanding how the axial forces on the DuraOmni rollers are reacted.

Can someone post attachments or links to

- assembly pictures

- assembly videos

- IPB

- CAD rendering of parts

They use white nylon on the inside of the roller as a bearing surface to handle axial and radial loads.

http://www.andymark.com/product-p/am-3048.htm

I believe other omni wheels use flanged bronze bushings.

-Adrian

[Ether]

Originally Posted by Adrian Clark

They use white nylon on the inside of the roller as a bearing surface to handle axial and radial loads.

When there's an axial load, what does that white nylon rub up against (material and shape) ?

[Ryan_Todd]

Interesting concept indeed! Looks like the large outer rollers would probably have to be molded in two halves, then bonded together. Not impossible, but also not particularly cheap. The real trouble will be with the hollowed-out portion of the larger rollers, because any material soft enough to be grippy will also be soft enough to flex inward and rub against the support inside it. One might perhaps skirt around this issue if the surface it would rub against is replaced with a bearing, however, which would be even more interesting to see.

I'll take your current design, and raise you an improvement:

Most of your assembly difficulties could be eliminated if, instead of alternating between small and large rollers, you made all of the rollers the same size, but asymmetrical: small on one end, and large on the other end. The small end of each roller would nest inside the large end of the next roller, and so on around the circumference. That way, you only have to solve the interference problem for a single roller instead of all of them, and your production costs would be much lower.

Here's a quick mock-up cross-sectional view of what I'd suggest:
(it's missing the roller pins, but you get the point)



The roller profiles and support arms would definitely need some tweaking to make them more structurally sound, and you'd need to work out the assembly process a bit more, but it's not all that far-fetched to think that this sort of thing could work IRL.

[GeeTwo]

Originally Posted by Skywalkar

I'll take your current design, and raise you an improvement.

I definitely like this one better. At first, I was worried about the variable distance from the pin to the roller, but then I realized that this is also a problem with the DuraOmni, or any omni with fewer than a couple dozen rollers. I can easily imagine how to have the wheel be together, but how it comes together is certainly a challenge. Perhaps if you had some spacers that entered through the hub and held the two halves of the wheel frame apart equal to the pin diameter, laid it horizontal, then maneuvered each wheel/pin in, and finally removed the spacers and bolted the two halves together? To do this properly with minimal roller wobble, you'd probably have to have the top and bottom halves not be reflection images.


Ether, as to Duraomnis: We used them this year for what was originally an H/slide drive. They are shipped assembled, and dis-assembly was not something obvious. I'll try to remember to take a look at ours on Saturday to see how they go together and if I can confidently take one apart find out what they're made of. By "axial load", I take it that you mean load parallel the roller axis, or equivalently, thrust/braking along the wheel's canonical direction of travel. Please advise if this is incorrect.

[Ether]

Originally Posted by GeeTwo

Duraomnis: ...I'll try to remember to take a look at ours on Saturday to see how they go together and if I can confidently take one apart find out what they're made of.

Thank you.

By "axial load", I take it that you mean load parallel the roller axis

Yes. The load that causes the roller to slide on its axle (one way or the other) until it hits something that stops it. What is the thing that it hits made of, and what is the shape and dimensions of the contact patch?

[BJC]

Another primary reason these types of Omni Wheels will not see use in FRC is due to the roller thickness. Both the original and the "improved" wheels have parts of the rollers with near zero thickness. This comes with a number of problems including that the rubber rollers preferred for traction won't support a robot's weight on near zero thickness (bumpy ride) and as the wheels wear they will very quickly become non-round (very bumpy ride.)

[Ryan_Todd]

Originally Posted by BJC

Both the original and the "improved" wheels have parts of the rollers with near zero thickness. This comes with a number of problems including that the rubber rollers preferred for traction won't support a robot's weight on near zero thickness (bumpy ride) and as the wheels wear they will very quickly become non-round (very bumpy ride.)

That's for sure! Neither of these designs could be expected to produce a perfectly smooth ride in real life, even after they were tweaked for durability and manufacturability. Even dualies can only be perfectly smooth if the wheel axle is held perfectly parallel to the ground (flat ground AND infinitely stiff robot frame), but with a good enough roller design, they can indeed usually be expected to produce a smoother ride than these two wheel designs would.

Engineering is all about trade-offs, however, and a design like these two would seem to strike a potentially useful balance between a traditional single-omni design and a dual-omni design: smoother than a single-omni, and lighter/more compact than a dual-omni. It still remains to be determined whether they could be produced at a low enough price point to compete with the current market leaders, of course!

-----------------------

EDIT:
Now there's an odd idea. I was considering refining my design and printing it in pieces just to evaluate the concept, but then I remembered that many "impossible-to-make" designs are trivial when additive manufacturing is employed. What if a multi-nozzle 3D printer were employed: nylon for the structure, TPE for the rollers, and a soluble support material to maintain clearance where necessary?

[Greg Needel]

you mean like this one?

[Adrian Clark]

Originally Posted by Ether

When there's an axial load, what does that white nylon rub up against (material and shape) ?

The white nylon rubs against the polycarbonte body of the wheel. Looking at the AM pdf it appears the roller contacts a flat face. I could be wrong on this, I'll leave it to someone with the time to open the cad to confirm.

-Adrian

[Ether]

Originally Posted by Adrian Clark

The white nylon rubs against the polycarbonte body of the wheel.

Polycarbonate. Hmm. Is there enough room between them to install a lower-friction (perhaps nylon or Teflon) shim?

[GeeTwo]

Originally Posted by Adrian Clark

The white nylon rubs against the polycarbonte body of the wheel.

Originally Posted by Ether

Polycarbonate. Hmm. Is there enough room between them to install a lower-friction (perhaps nylon or Teflon) shim?

It would have to more like a film than a shim; the free axial motion of the roller is barely sensible. If you were willing to trim down (and presumably polish) the roller you could put a thin shim in place.

[Ether]

I couldn't find any data for polycarb-on-nylon friction. I suspect it's not pretty.

[GeeTwo]

Originally Posted by Ether

I couldn't find any data for polycarb-on-nylon friction. I suspect it's not pretty.

I concur. However, in many cases where omnis are used, the goal is either to allow a long wheelbase robot to turn (tank with 2 or 4 omnis), or to allow fine strafing adjustments so that a robot can line up on a game piece or a target after forward motion has completed or nearly completed (H/slide drive). In both of these cases, "not pretty" is still a whole lot better than "solid wheels".

I can certainly see roller friction while applying driving torque being a real problem with kiwi (three omni wheels in a triangle) or Killough (four wheels in a diamond configuration) drives. I almost suggested kiwi (implicitly using duraomnis due to our game's KoP) to one of our intramural design teams today; I'm glad I held back.

[T^2]

Originally Posted by Ether

I couldn't find any data for polycarb-on-nylon friction. I suspect it's not pretty.

Nationwide Plastics has polycarb-polycarb at 0.52 (unspecified, but I'd guess static) and Engineering Toolbox has nylon-nylon at 0.15-0.25 (static). Polycarb-nylon might be somewhere in between. Not too bad, really.

[pmckinnon]

The concept of these wheels are quite good as they provide close to a full concentric circumference, however they are complex to manufacture. A greater choice of dual and triple row omni-wheels are more readily available such as those by Vex, Andymark and Rotacaster Wheel which can provide better traction and better or equivalent ride quality.