Variable Diameter Wheels

[KenWittlief]

depending on where the axle pivots, the robot can be designed to be higher on the larger wheels, or higher on the smaller wheels, or it can be exactly the same height throughout its CVT range

it all depends on the shape of the wheel, and where the pivot point is.

[sciguy125]

Let's think about this for a second:

Let's make a sphere. Mount a motor, gearbox, and whatever else inside of this sphere. Drive a shaft throw the center of the sphere, perpendicular to the motor shaft. Mount the assembly to the robot using this new shaft. Remove any excess portion of the sphere that will not contact the ground during normal operation - the sphere has become a wheel. Now, you need some way of pivoting the assembly about the mounting shaft.

This configuration will make it so that the height of the chassis never changes. However, you can change this by moving the mounting shaft closer or farther away from the robot (relative to the assembly).

[KenWittlief]

Originally Posted by sciguy125

This configuration will make it so that the height of the chassis never changes. However, you can change this by moving the mounting shaft closer or farther away from the robot (relative to the assembly).

yes, relative to the center of the sphere

or if you enlongate the sphere the height of the robot can also be different, depending on where the pivot point is.

[Madison]

For what it's worth, the sketch I posted earlier in this thread is, more or less, exactly as Phil has described. The notable difference is that I presumed that the wheels would not be powered independently.

They pivot, however, such that the axis of rotation passes through the center of the sphere, thus only the effective diameter of the wheel changes and never its position relative to the chassis.

[Cuog]

Actually looking at your drawing it apears to me that since the pivot point is slightly behind the center it would then lift and lower the robot however i am estimating this difference is about an inch and not a large problem in most competitions(unless they force us to fit under something)

[Madison]

Originally Posted by Cuog

Actually looking at your drawing it apears to me that since the pivot point is slightly behind the center it would then lift and lower the robot however i am estimating this difference is about an inch and not a large problem in most competitions(unless they force us to fit under something)

The pivot is not slightly behind the center; it passes through the center. The 'wheel' is not a full hemisphere.

[JVN]

As you move the wheel from "fast" mode to "torque" mode, do you believe the frictional load caused by sliding the sphere sideways on the carpet will cause issue?

My goal would be to make these as high traction as possible, to get the use out of my torque in the low end. I'm worried that this high traction would seriously impede the movement from high-low and vice versa.

We're talking about a not insignificant amount of force.
Just something to consider.

This is definitely an interesting concept. I wish I had more time to build a proof-of-concept out of Vex. I hope this gets further developed, and possibly put into practical application.


-JV

[Madison]

Originally Posted by JVN

As you move the wheel from "fast" mode to "torque" mode, do you believe the frictional load caused by sliding the sphere sideways on the carpet will cause issue?

My goal would be to make these as high traction as possible, to get the use out of my torque in the low end. I'm worried that this high traction would seriously impede the movement from high-low and vice versa.

We're talking about a not insignificant amount of force.
Just something to consider.

This is definitely an interesting concept. I wish I had more time to build a proof-of-concept out of Vex. I hope this gets further developed, and possibly put into practical application.


-JV

The force required to rotate the sphere would be about the same as that required to move the robot since the rotation is, in effect, acting as little more than another wheel. While moving the robot certainly requires a decent amount of force, it is often not the most difficult task to accomplish on a robot and, given that this rotation need only be ~90*, it is likely possible to accomplish it using some method that, while powerful, is not suited to drivetrains.

I don't think anyone's really done any considerable calculations regarding this; at least not here. I certainly have not. I just tossed the model together as a visual representation of the concept. I have my hands full with something else just now, so I won't be trying to build something like this for a long time to come.

Edit: This reads as if I'm being preachy -- I'm not, since I know you this stuff. I was just being expository.

[Alan Anderson]

Originally Posted by JVN

As you move the wheel from "fast" mode to "torque" mode, do you believe the frictional load caused by sliding the sphere sideways on the carpet will cause issue?

My goal would be to make these as high traction as possible, to get the use out of my torque in the low end. I'm worried that this high traction would seriously impede the movement from high-low and vice versa.

Nobody said the sideways rotation has to be symmetric around the center of the robot. If the hemispheres are all turned in the same direction, the whole robot will shift a bit to one side, and there won't be any carpet friction problem.

[Cuog]

Well my team has a meeting today after school so I'll probably bring it up and see if we can build a test component to see how well it works

[greencactus3]

well if we have stairs again it might be useful. two birds with one stone.
the changing rideheight i mean.

[Jay TenBrink]

[quote=JVN]As you move the wheel from "fast" mode to "torque" mode, do you believe the frictional load caused by sliding the sphere sideways on the carpet will cause issue?

John,

I think the bigger challenge to overcome would be the losses from scuffing while going straight. The contact patch where the tire meets the carpet would approximate a section of a cone. This conic section will want to turn about a point at the apex of this cone. Just think of how a styrofoam cup rolls in a circle. Car tires will do the same thing and can cause your vehicle to pull yo one side if they have excessive conicity, which is one side larger than the other.

Assuming the conicity of the tires would all point in (or out), the forces would cancel each other out. There would still be high rolling resistance to overcome.
Sounds like an interesting idea.

Jay