HDPE or ICE FLOOR

[samir13k]

Originally Posted by dtengineering

Okay... I really like the saw blade idea, and will always cast my ballot in Q's camp. But I doubt they'd make it through tech.

So although my fellow Canadians may strip me of my citizenship for letting out this little piece of information, nothing grips ice quite like broomball shoes. Bet you didn't even know broomball was a sport. (Well, aside from those of you in the other snowy bits of the world.) Mind you, from my experience a key aspect of broomball involves beverages that would be out of place at an FRC competition.... so my impression of the outstanding grip of broomball shoes may be... um... imprecise... at best.

Broomball... Like Curling?

Also, take some thick sheet metal, curv it around kitbot wheels, and sharpen them so that they are like ice blade "blades". Maybe even for braking you can make them slide sideways and kick up all that ice...

....except they would prolly go with HDPE

[gorrilla]

conveyor belts for traction,

luckily we are only miles from g&t conveyor systems (they make airline baggage handeling things)

[Pat McCarthy]

We certainly know the idea has been floating around in the GDC's heads for a few years:


http://www.chiefdelphi.com/media/photos/22422

[Cory]

Originally Posted by gorrilla

conveyor belts for traction,

luckily we are only miles from g&t conveyor systems (they make airline baggage handeling things)

Like I said earlier, surface area doesn't matter.

HDPE and ice are as close as possible to an "ideal" surface from a physics standpoint. The reason surface area might matter on carpet is because the tread can interlock with the carpet.

On ice or HDPE the surface is totally slick, so there is no possible way for the tread to interlock with the HDPE, and thus surface area does not matter, only coefficient of friction and normal force.

[maltz1881]

HDPE is what milk jugs are made of. We have made several robots out of HDPE. Very easy to work with and we keep lots of it on hand. Very strong and durable.

[Gdeaver]

For the stack attack game, we found that McMaster-Carr Grip all neoprene rubber had very good traction on HDPE. Our problem was that it gripped to well on carpet (4" wide wheel). If we had less width and more power than the drill motors it would have worked well.

[gorrilla]

Originally Posted by Cory

Like I said earlier, surface area doesn't matter.

HDPE and ice are as close as possible to an "ideal" surface from a physics standpoint. The reason surface area might matter on carpet is because the tread can interlock with the carpet.

On ice or HDPE the surface is totally slick, so there is no possible way for the tread to interlock with the HDPE, and thus surface area does not matter, only coefficient of friction and normal force.

the conveyor belts they make are for moving bags around an airport, they are a like tiny little plastic plates with rubber tread on them, were you thinking like timing belt kinda thing?

[EricH]

Originally Posted by gorrilla

the conveyor belts they make are for moving bags around an airport, they are a like tiny little plastic plates with rubber tread on them, were you thinking like timing belt kinda thing?

Traction is directly related to frictional force. Force = mu * N, where mu is the coefficient of friction and N is the normal force (force directed directly away from a surface that something with friction is on). Notice the lack of area in that equation? Yep, there's a reason for that. Namely, surface area doesn't matter.

There's been discussion of this issue already, some years ago. Look it up.

And, the only way to increase mu is to increase the coefficient of friction. You can a) change materials or b) dig one material into the other. If you try doing b) to an HDPE section of the field, I'll call a rules violation, assuming there is a "no damaging the field" rule again.

[gorrilla]

Originally Posted by EricH

Traction is directly related to frictional force. Force = mu * N, where mu is the coefficient of friction and N is the normal force (force directed directly away from a surface that something with friction is on). Notice the lack of area in that equation? Yep, there's a reason for that. Namely, surface area doesn't matter.

There's been discussion of this issue already, some years ago. Look it up.

And, the only way to increase mu is to increase the coefficient of friction. You can a) change materials or b) dig one material into the other. If you try doing b) to an HDPE section of the field, I'll call a rules violation, assuming there is a "no damaging the field" rule again.

i know, i think ive read that......

i was thinking treads becuase i felt they would give the most stable platform on a slick surface

really its either that or wheels, and i dont think they would make the entire floor HDPE.........

[EricH]

Originally Posted by gorrilla

i know, i think ive read that......

i was thinking treads becuase i felt they would give the most stable platform on a slick surface

really its either that or wheels, and i dont think they would make the entire floor HDPE.........

Ah, stability? Again, not determined by surface area. It's determined by some other things, like CG and wheelbase.

Let's say you have a rectangular robot at max dimensions, with the wheels (or treads) a little bit inside the max. It looks like this:
_________________
||.....................||
||.....................||
||.....................||
||.....................||
||.....................||
||.....................||
||.....................||
||.....................||
||.....................||
||.....................||

That's the wheelbase as seen from the top. As long as you keep your CG above that area, you won't tip. As soon as the CG goes outside that area, you've got a split second to get it back before the big crash. If you have treads, you'll still have the same area.

[gorrilla]

Originally Posted by EricH

Ah, stability? Again, not determined by surface area. It's determined by some other things, like CG and wheelbase.

Let's say you have a rectangular robot at max dimensions, with the wheels (or treads) a little bit inside the max. It looks like this:
_________________
||.....................||
||.....................||
||.....................||
||.....................||
||.....................||
||.....................||
||.....................||
||.....................||
||.....................||
||.....................||

That's the wheelbase as seen from the top. As long as you keep your CG above that area, you won't tip. As soon as the CG goes outside that area, you've got a split second to get it back before the big crash. If you have treads, you'll still have the same area.

ah, wheelbase(couldent think of the word for it) thats what i meant......

i never implied that they were going to be wider than a wheel..............

there like this
http://www.chiefdelphi.com/media/photos/28106

except flat

[gblake]

Originally Posted by EricH

Ah, stability? Again, not determined by surface area. It's determined by some other things, like CG and wheelbase.

Let's say you have a rectangular robot at max dimensions, with the wheels (or treads) a little bit inside the max. It looks like this:
_________________
||.....................||
||.....................||
||.....................||
||.....................||
||.....................||
||.....................||
||.....................||
||.....................||
||.....................||
||.....................||

That's the wheelbase as seen from the top. As long as you keep your CG above that area, you won't tip. As soon as the CG goes outside that area, you've got a split second to get it back before the big crash. If you have treads, you'll still have the same area.

Erich - Don't forget to remind your audience that inertia exists, and that it makes a moving robot do things that a stationary robot would not (like tip over even if the CoG is within its wheelbase)...

[EricH]

Originally Posted by gblake

Erich - Don't forget to remind your audience that inertia exists, and that it makes a moving robot do things that a stationary robot would not (like tip over even if the CoG is within its wheelbase)...

Oh, Newton's first law?

That is true, especially if there's a sudden stop or change of angle involved. See, in physics (and in Statics), you can usually get a really, really good approximation of forces and how they'll react with each other by finding the CG and having all the forces act from that, unless you're dead sure that the force acts elsewhere (like friction, which gets as close as it can to the floor, or an impact from another robot, which should be within the bumper zone).

If you've got a robot traveling at constant speed, the only forces acting on it are weight, the normal force (cancels out the weight), friction (applied to the wheels) and whatever force you need to maintain equilibrium (applied to the wheels). Now, a sudden stop! The force needed to maintain equilibrium has been effectively distributed through the robot, and there is a sudden forward force at the CG (for simplicity--there's a rule saying I can move it there, I just forget which one) and a reverse force on the lower section of the frame. This creates a moment, causing the robot to rotate forwards. If a corrective force is not applied, the robot continues to rotate and goes splat. This effect is not helped if the CG is high, as there is now a much bigger "lever" for the forward force to act on, and it takes much more force to stop due to the way moment is calculated.

[Bongle]

Originally Posted by EricH

Namely, surface area doesn't matter.

This isn't _entirely_ true. In an ideal physics world, yes, but there are cases where surface area is proportional (or inversely proportional) to grip.

-Racing cars have very large, wide tires (even though this increases unsprung weight) because it means that if the tire hits an imperfection in the track, it doesn't lose grip. Since the track is always imperfect, this has the effect of effectively very slightly raising the car's coefficient of friction (although the ideal rubber-on-ashphalt coefficient remains the same)
-Skates have very little surface area because it takes a lot of pressure at the skate-ice interface to create the microscopic layer of ice that the skate glides on.

Likewise, I'm not sure the rubber-on-carpet case is a cut-and-dry Ff = uFn case. It might be like Velcro: if you have two big sheets together, it is much harder to pull one off sideways than if you have two small sheets together. Despite the normal force being the same, it takes much more force to move the two sheets relative to each other when there is more surface area in contact.

There have probably been some teams that have done tests, it would be interesting to see the results.

[JesseK]

If they do a mostly-HDPE field, I think we'll see alot more tank tread attempts for drive trains this year. The successful ones will have the right tread pattern.

However, I suspect that if we see any HDPE, it will only be on part of the field, or a small field element. That stuff isn't exactly the cheapest material around...