Variable Diameter Wheels

[EricH]

Originally Posted by Adam Y.

Believe it or not the wheels on a train are cone shaped even though you can't tell by staring at them. I think they are shaped like that because otherwise the train would derail easily.

Physics--There's a reason for it (to keep the train on the track, in this case). I forget the name of the principle. Here's a quick demo of how it works: attach the bottoms of two styrofoam cups (you know, those little white ones that come in huge packages) together, set up two rulers/metersticks (or something similar) parallel to each other and roll the cups down from one end to the other. That's what keeps the trains on the track (the flanges are backup).

[KenWittlief]

Originally Posted by EricH

Physics--There's a reason for it (to keep the train on the track, in this case). I forget the name of the principle. Here's a quick demo of how it works: attach the bottoms of two styrofoam cups (you know, those little white ones that come in huge packages) together, set up two rulers/metersticks (or something similar) parallel to each other and roll the cups down from one end to the other. That's what keeps the trains on the track (the flanges are backup).

its more that that. The train wheels are welded to the same axle, so as the train goes around a bend the centripetal force causes the outside wheel to ride up on its inside (larger diameter) and the wheel on the inside of the curve is on the smaller part of its diameter. This allows the wheels to stay in contact with the rails without slipping, because the outside rail is longer than the inside rail on a curve.

[dlavery]

Originally Posted by KenWittlief

its more that that. The train wheels are welded to the same axle, so as the train goes around a bend the centripetal force causes the outside wheel to ride up on its inside (larger diameter) and the wheel on the inside of the curve is on the smaller part of its diameter. This allows the wheels to stay in contact with the rails without slipping, because the outside rail is longer than the inside rail on a curve.

Just to finish off this train of thought (ha!), the AAR (Amercian Association of Railroads) standards established a wheel profile with a 3 degree slope across the width of the wheel tread. These standards were later adopted by the Federal Railway Administration. The fillet between the wheel tread and the flange used to be a simple circular radius. But in the past thirty years, as metallurgy has advanced and resulted in harder railhead materials, this simple profile caused uneven wheel wear and hollowing. In response, the fillet proifle is now a convoluted shape that minimizes wheel wear on the new, harder rails.

...and now back to our originally scheduled topic of discussion.

-dave

[JVN]

Originally Posted by dlavery

...and now back to our originally scheduled topic of discussion.

-dave

Quit railroading everyone. We need to keep this thread on track.

Har Har Har... Quite punny.

So anywhoo, The only real problem that I see with this from a FIRSt bot perspective is weight. The wheels would would have to be considerably heavier than normal, as well as adding a whole mechanism for tilting the wheels controllably, while retaining some structural stablity. Anyone got any ideas on how to keep the weight down (besides cheeseholing) ?

[KenWittlief]

I was thinking instead of making it infinately variable, you could use the same concept and have maybe 3 wheels (three speeds).

This would also allow you to make the smaller diameter wheels have wider treads, for more traction, and have the taller wheels wearing skinny tires.

and you could angle each of the three wheels to keep the tread parallel to the floor when it is rotated into positionl

[greencactus3]

Originally Posted by 114ManualLabor

Har Har Har... Quite punny.

So anywhoo, The only real problem that I see with this from a FIRSt bot perspective is weight. The wheels would would have to be considerably heavier than normal, as well as adding a whole mechanism for tilting the wheels controllably, while retaining some structural stablity. Anyone got any ideas on how to keep the weight down (besides cheeseholing) ?

but then again you dont have a couple gears that would be needed to gain the gear shifts. and well other than a cvt, its probably one of the smoothest gear transitions possible. and infinite.

[Billfred]

Originally Posted by KenWittlief

I was thinking instead of making it infinately variable, you could use the same concept and have maybe 3 wheels (three speeds).

This would also allow you to make the smaller diameter wheels have wider treads, for more traction, and have the taller wheels wearing skinny tires.

and you could angle each of the three wheels to keep the tread parallel to the floor when it is rotated into positionl

Out of curiosity, how would you fabricate such a machine?

[greencactus3]

Originally Posted by Billfred

Out of curiosity, how would you fabricate such a machine?

i woulda say lathe. not too hard to shape flat surfaces on an angle. just need a protractor and a chunk of wod and a ruler. well or a CNC

[Cuog]

The 3 wheels idea sounds scary to fabricate since you cant take a rod and stick 3 wheels on it the rod would have to constantly bend so that only 1 wheel would contact the ground at a time or the relative wheel diameters would have to be carfuly selected and placed with very exact positioning

[ahecht]

Well, if you are talking about variable diameter wheels, but now continuously variable diameter wheels, 190's 2001 robot did use two sets of different sized wheels to shift on the fly (one set could be raised or lowered). There has also been a clever design for a continuously variable wheelplan we've been kicking around for as long as I've been on the team, but I don't think any of the recent games have really made it necessary to use.

[Jeffrafa]

Believe it or not, i used to have an RC truck that worked just like you're describing. I actually only realized how it worked after reading this (given I havn't played with it in years - I knew it went slower when stood up, but i never had thought about why).

It had the motors built into the rotating part with the wheels, as you've discussed. It was skid steer, and it would stand up or squat down - giving it the adjustable speed.

I've found one picture online. I'll see if i can find mine in the garage tomorrow to get better pictures. We moved last summer though, so I dunno how lucky i'll be.




- Jeff

[greencactus3]

Originally Posted by Jeffrafa

- Jeff

yup. thats what i was talking about.. such an ugly rc truck tho lol.
obvously that toy's rideheight changes a considerable amount. have we figured olut if thats good or not in our robots?

[Gerry Salinas]

Originally Posted by greencactus3

yup. thats what i was talking about.. such an ugly rc truck tho lol.
obvously that toy's rideheight changes a considerable amount. have we figured olut if thats good or not in our robots?

I'd say it's almost certainly not a good thing, for several reasons; it's easy for the car to lift itself up, but a 120 lb. robot will have a harder time supporting its own weight, I should think. Beyond that it can depend on the game and the robot, but changing the height of the robot certainly makes it harder to be certain of where anything is for autonomous or even just operation, although I doubt this would be as much of an issue as the weight thing.

[Greg Needel]

Originally Posted by greencactus3

yup. thats what i was talking about.. such an ugly rc truck tho lol.
obvously that toy's rideheight changes a considerable amount. have we figured olut if thats good or not in our robots?

OK i will finally come clean that i have been working on this system for quite a long time now, you will see it sometime soon, once i get it completely finished. As for ride height i personally wouldn't want my CG to change (to be at it's highest) when i am getting maximum torque.