I’ve searched and I really didn’t find any discussions that directly answer these questions. If you know of a posting or white paper that addresses these questions, can you please post a link to it, otherwise, what are your thoughts, and why?
I’ve read through several white papers, but none seem to address these specific questions.
If gearing was appropriately matched, what are the advantages and disadvantages of the various diameters of wheels? Does the width of the wheel really matter that much?
EDIT: This is as much for general consumption as it is for my education.
As a general thing, obstacle traversing is better with a larger wheel (larger wheel, larger obstacle). However, larger wheels can easily be more weight, and a smaller wheel can be made to climb as effectively as a large one. Pick your poison; I’m not quite sure whether there’s enough difference between a 6" and 8" or between a 4" and a 6" to make it worthwhile. You only get about an extra inch of vertical climbing ability. 4" to 8", probably enough to make it worthwhile.
On width, ideal physics says no difference, while reality says wider is better. I’ll let the idealists and the realists argue that one out, preferably by testing.
Gearing matched: Given that the gearing is matched to give the same robot speed, a smaller wheel will tend to need a smaller reduction (less weight) to go as fast as a larger wheel. (Smaller reduction = higher rotational speed, for those that might be confused.)
Cost, tread life, ability to go over obstacles, ground clearance, room for sprockets/chains, and robot stability (you can have a longer wheelbase with smaller wheels), are a few of the considerations that I can think of.
OK, so smaller wheels have a distinct advantage for lowering the CG of the robot and maybe reducing the weight a bit, but also reduce ground clearance if it is needed. Larger wheels allow you to traverse a larger step change in the driving surface, but not much advantage on a ramp change.
I think I can help you out a little. The width of a wheel does matter. A 1" wide wheel has half the contact patch of a 2" wide wheel. ex: You don’t ever see dragsters with thin wheels. They always have very thick wheels giving them a higher normal force so they can take off faster. That said, with FIRST robots it is a pretty small difference. If you are looking to increase your traction, you should look for wheels with a higher coeffient of friction. As far as I know pneumatic wheels have the highest CoF (about 2.0). Other wheels have more but are not legal in FIRST because they destroy the carpet.
As to wheel size there are lots of pros and cons to each. Weight, speed, efficiency to name a few. It’s really personal preference, do you need the weight? do you want low/high ground clearence?, ect.
As to whitepapers I think one team (234?) is working on testing various sized wheels for differences in preformance. However, I wouldn’t expect it to be out any time soon.
Bryan, wider wheel does not give higher normal force because it is a wider wheel. Normal force is the same–object mass/# of object points of contact.
However, tires are rubber. Rubber, when warmed to a certain point, gets “stickier”. In that case, more surface area contact at a contact point is better.
If you’re going to assert reality contradicts physics, please at least cite a source of test results or other evidence. Statements of fact without backing like this are how misconceptions are made.
Anyway, there are a lot of reasons to like smaller wheels. The wheels themselves weigh less, and less of a reduction is needed to achieve the same output properties as a larger wheel. The primary drawback is that smaller wheels require more effort to climb obstacles with - a smaller wheel inherently has lower potential ground clearance than a larger wheel. With proper design this can be worked around - many teams climbed the bump this year with 4" wheels and the bump was a rather aggressive obstacle.
There is a team working on testing this. As such, there are no definitive results yet. It’s quite possible that theory and reality agree. But there have been some CD discussions on this topic over the years, which taken together indicate that there is some disparity, due mostly to rubber interacting closely with carpet to give a slight but noticeable increase in traction on a wider wheel.
Remember: Ideal physics tend to take place on a surface with uniform friction (possibly no friction) and in a vacuum unless otherwise specified. There are very few places that have both. Because that sort of place is rare, especially when Murphy is around (or at a robotics competition), reality tends to win over physics by emphasizing those slight differences.
I also distinctly recall Andy Baker making some comments on this. Where they did some tests and found that a wider tread DOES have more grip on carpet.
Suffice to say this debate has been going on for as long as most of us have been around these boards.
Wider tread provides absolutely no benefit. When the surfaces are hard, uniform, and flat (Lunacy)
A wider tread does increase the amount of gripping ability when you are talking about carpet though. Think of carpet wheels kinda like Velcro, when you drive a carpet wheel over carpet, some amount of interweaving occurs, resulting in some amount of grip based on contact area.
Whether or not the amount of increased grip is usable is still up in the air, but I am betting that it would increase grip a small amount.
Unfortunately, in the case of polymers, this is not a misconception. On a microscopic level, polymer-based materials (such as the rubber on wheel treads) have “knobs” that protrude and grab on to the microscopic “hooks” of fibrous materials (such as carpet). For this reason, increasing the contact pad increases the number of “knob” to “hook” interactions, which in turn provides an increase of traction. Idealistic entry-level physics states that F=uN, and so as long as the normal force and coefficient of friction remain the same, (it is commonplace to assume that surface area has no affect on the coefficient for basic physics because the other equations are far to complex for classroom labs) no change in frictional force will occur, regardless of the surface area.
Think of it like Velcro. What’s harder to pull straight off, one hook grabbing on to one loop, or one inch of hooks on one inch of loops.
“F= uN” is physics
To build on what Ether said, this is only a rough approximation of the resistant force felt by an object, and lumps together everything from microscopic surface roughness to how deep one object sinks into another (e.g. a marble on carpet). There are, however, many other equations that these factors into account.
Sorry if this repeats any of the information on the linked threads, it’s just nice to have all the important information in one place.
Do any of you have measured data of how this phenomena applies to an FRC application? Anyone?
Do any of you whom are convinced about this fact with long explanations know how much surface area comes into play? Numbers and rates, not “a little” or “a lot”.
Wider wheels (or tank treads) = larger contact patch. Larger contact patch = more grip than smaller contact patch. More grip = more traction for the same surface. Therefore, more drivetrain contact patch = more traction, under carpet conditions.
Yes, I understand that more = more, that’s not what I was asking at all.
My point was that accepting a statement as truth without understanding the magnitude of the effect, or knowing how much the effect applies, is not a good idea. Especially when “conventional wisdom” has all of those things.
It seems like a lot of people are parroting back stuff they heard from some guy instead of relying on observations and data. How is that good engineering?
For roughtop tread, I am going to make an assumption that smaller diameter wheels get more grip as they form more of a sharp cleat than larger diameter wheels do. This cleat is a better shape for interlocking with the rows of carpet fibers.
Making it wider still retains this shape, but increases the width of the cleat.
This is purely based on the anecdotal evidence of our 2008 krab drive running 2" wide wheels.
You are correct of course, doing the experiment yourself/having well documented reports is the proper way. I will try to find the thread I recall reading it in, it was from a few years ago. (I want to say 2003)
Maybe, when first-hand observations are scarce, you have to go by hearsay or not go at all. Have you ever actually observed that F=ma? Have you ever measured the coefficient of friction without using F=uN? How do you know, then, that what you’ve heard is correct? Maybe F=ma^0.9999999999999993, or F=u^0.999999995N measured experimentally. *Every time that we as engineers use an equation we’ve learned, we’re parroting back stuff we heard from some guy. *
The only difference is that for these established equations, “some guy” is whoever wrote the textbook and whoever taught the classes. Sure, they’ve been backed up by decades or centuries of testing. But only a few tests have actually been run on wheel width vs. traction at the FRC level. I can only think of 2, and only 1 has actually been finished and put out there so far.
