66 needs your help!

[David66]

we began our drivetrain with pnuematic 12.5 tires. Then our robot began bouncing after a very tall addition. So we gusseted the tall addition, but to no avain. SO we decided to attempt to use solid wheels. This quited the bouncing (while turning) down a bit, however, it is far from gone. The robot will spin very nicely on the HDPE meaning we have a friction problem...whathave you done (if you have a tall tower on your robots) to prevent your robot fromdoing this waddle/skip/dance/bounce? any input wuold be greatly appriciated

[Rob Colatutto]

You could try slightly lessening the pressure in the pneumatic tires. Keep in mind the less pressure in them the more contact you will have with the ground and thus more wheel to drag when turning. While playing with the pressure in pneumatic wheels once I noticed that with lower pressures the wheels do bounce much less when dropped so that could be a quick solution.

[David66]

tried that...but ideas are good...lets keep em coming until we hit one something

[Dick Linn]

Originally Posted by David66

tried that...but ideas are good...lets keep em coming until we hit one something

Using 4 wheel drive? If so, is it the alternate grabbing and slipping (stick/slip) in turns that is contributing to this motion? If you're 4 WD, maybe a slightly less stick tire compound on one end would help. If you have a tall appendage (lever arm) with a significant amount of weight up high you may have to rethink things, or adjust your driving style to slow down on the turns.

[Kevin Sevcik]

The problem is that you're trying to force the wheels to slide sideways because you're using 4 wheel drive. I'm guessing your effective wheelbase is longer in the forward backward direction than in the left right directions. If you think about it, this forces the wheels to slide side to side.
To use an extreme example, imagine trying to make a bicycle turn in place. The wheels have to slip side to side and their turning forwards doesn't help at all.
Now imagine a segway, it turns on a dime with no problem, and the wheels don't have to slip slide to side.

So you have a few solutions, but this late in the game you have to decide what's practical:
1. go two wheel drive with casters or skid plates. (I;m not trying to start the usual holy way about 2 vs. 4-wheel, I swear).
2. make your wheelbase more segway-like by moving the wheels closer together so the wheelbase is wider than it is narrow. Team 57 did this this year and our robot turns MUCH better.
3. bald one pair of tires so it will slip side to side.
4. go 6-wheel drive. the middle pair will force the robot to turn since it will be applying force directly to turn the robot like the segway wheels.
5. Make Omni-Wheels for one pair of wheels.

Keep in mind this is just from my experience with my team's robot, though. I've seen other teams make 4-WD robots that turn with no problems, and I've no clue how they did it.

[edomus]

Omni Wheels

[sanddrag]

I don't know about the legality of this (since I believe it would be a hard plastic tractive surface) but you can try putting zip ties through your wheels and around your tires. Let me know if you don't understand what I mean and I'll post a pic.

Also, try airing up one set and airing down the other set.

[WakeZero]

Bouncing how? If it is bouncing in the way I am thinking, it may be because you are accelerating too fast. You could try and fix this in programming by making your drive motors gradually increase to full speed?

[Al Skierkiewicz]

As other have stated, your robot is reacting to the friction developed between the four wheels and the carpet in turns. You have to either turn two of the wheels when turning or reduce the side friction while turning. There are several ways to do this, one is to lift two wheels off the floor while turning. Another is to use a special type of wheel where there are rollers on the edge. It grabs in the forward motion but the rollers turn when the wheel is moving sideways. We developed such a wheel and the picutre is in the archive. Our mech engineer, Raul, calls them BUPOD wheels and they worked really great.

[Alexander McGee]

If it is 4x4, a quick, very good answer is pop-down casters on one side of the robot (front or back). You can either use one or two, and easilly position them with pnumatic cylinders.

[Dick Linn]

Originally Posted by magnasmific

If it is 4x4, a quick, very good answer is pop-down casters on one side of the robot (front or back). You can either use one or two, and easilly position them with pnumatic cylinders.

Yes, that or just a single big "hockey puck" made out of HDPE in the middle, beveled on the bottom so as not to snag. Or a bar of HDPE hinged on one end, whatever is simple and quick. Casters can handle funny, so we're gravitating away from them.

4-wheelers should have a short wheelbase and wide track in order to balance the leverage forces. Last year ot bot had about a 17 inch wheelbase and was fairly narrow, so we ended up using 2 pneumatic Skyway wheels and two of the solid polyurethane wheels (that were in the kit) to make turning a bit easier. It's still a bear to turn on pavement, but not too bad on carpet.

[Schaffer]

What you might want to do is move your two drive wheels closer together (for a 12" wheel - 18" axel to axel is good as long as it is far enough away from the CG in the vehicles axel plain) and put a castered wheel out where the old drive wheel was. This way you perserve the drive train integrity and stop the bouncing...

[KenWittlief]

Im not sure if this will help any - last year we had a large alum plate in the center of the upper part of our robot, that had pneumatic cylinders attached to it - it went all the way across the 'upper frame'

we changed our bot and took that plate off, and I noticed in our next match our bot was swaying back and forth, looked like it was drunk

back in the pitts I realized that plate had been the only source of diagonal rigidity in the upper frame - always gotta have triangles in any square/boxy type of framework to make the structure stiff and strong

BTW - I also noticed the staggering when the bot was turning - and it was also 4WD.

maybe you need to stiffing up your framework - put diagonal cross members, or reinforce the corners with triangles. If you are planning on putting plastic side covers on, that would stiffin the frame up too.

Only other thing I would say is maybe your center of gravity is too high and you have created a Weeble Bot - can you move your battery lower, or put it horizontal? or move anything else down to lower your center of gravity?

but in the end, if the bot looks like its drunk when it turns, does it really matter? If it works, it works - and if you cant fix it your bot will have a unique charactistic that will lend itself to many names that I can think of off the top of my head:

Otis (from the andy griffin show)
Weeble-Bot (weeble wobbles but he dont fall down)
Sir Osis of Liver
Rocky
.....

:c)

[ahecht]

The real key, if using 4 wheel drive, is to have a pair of high traction wheels closest to the center of gravity, and a pair of lot traction wheels on the other side.

For example, you can turn down your skyway wheels on a lathe, one pair just until the first layer of plastic shows, and the other pair 1/8" or so beyond that. Then epoxy and screw a thin strip of incline conveyor belt material (very high friction) to the smaller pair. You should now have two pairs of wheels, about the same size, with vastly different amounts of grip. This solution was used by 992 when I was on it in 2002, and by 190 in 2003 (although we originally had conveyor belting on our front wheels, but it came off in competition). Both bots also had a very short length and a wide wheelbase, which helped.

On the other hand, 190's current robot is very long with a narrow wheelbase, and all our tires our pneumatic. Needless to say, turning won't be a major part of our strategy this year.

[Dick Linn]

Last season, we used 2 of the pneumatic Skyway wheels on one end, then a pair of the poly wheels on the other (8.75 or so inch diameter). Since the poly wheels were not the exact diameter of the rubber wheels, we "turned them down" by holding them between our knees and using a belt sander and a bit of elbow grease (not to mention finesse) to get them even and slightly radiused in the axial direction. Talk about a mess of black dust! Glad we did it at my employer's shop and not at school

Because we were climbing a steel mesh that year, we re-grooved the original channels across the tire using a 1/4 inch rat-tail wood file. The result was a bot that could climb the wire ramp (with a 110 lb. person on board) but still manage to turn, since the poly wheels were just a bit more slippery than the rubber ones. The axial grooves bit into the mesh on the way up to provide traction. This year the best way depends on your method of attacking the platform and whether or not you see it as a pushing/pulling contest.

As to turning and handling, you either play around with the geometry, center of gravity and traction, or gear it down to be a real brute to get the desired results. Adjusting for handling can be both a science and an art. At this point in the build, I think it depends on how much time you have left. Quickly try it on a different surface to see if things change dramatically. You might discover something.