Lowered Center Wheel for 6 wheel drive

[Josh Murphy]

we have used 6 wheel drive on 2 of our last 3 and I know that this year it was lowered 1/8 and in 04' I am pretty sure it was the same.

[Mark McLeod]

Originally Posted by Paul Copioli

We will probably go with 3m this year.

3 meters seems a bit extreme

[Josh Murphy]

Originally Posted by Mark McLeod

3 meters seems a bit extreme

I am sure that he ment 3mm if you read the post. .

[Lil' Lavery]

Originally Posted by Paul Copioli

We will probably go with 3m this year.

I take it you have the inside scoop that the height restriction is going to be quite tall in 2007?

[Andrew Blair]

Originally Posted by Lil' Lavery

I take it you have the inside scoop that the height restriction is going to be quite tall in 2007?

No, he's talking about new tread material... I guess FIRST is lifting the "No Adhesives" ban... ;-)

[Wayne C.]

you don't lower the middle wheel-

you use progressively larger tires from front to back

[Billfred]

Originally Posted by Wayne C.

you don't lower the middle wheel-

you use progressively larger tires from front to back

<offtopic>Can someone explain in businessmajorese how this would work effectively? It sounds interesting.</offtopic>

[b-rant]

I believe 1/8" is the best option but if you're not sure I would definitely stick with getting pneumatic wheels for your center wheels so you can adjust the wobble with air pressure. Check out skyway for some nice wheels.

[nparikh]

We've been debating using a Nascar technique to robot design. After having realized that the most efficient setup for a 6 wheel robot is to lower the center wheel, we statistically analyzed all of our matches and noted that we turn left much more than we turn right, and as such we are considering making the wheels on the right hand side a tad bit thicker (perhaps a few mm), and having the left hand train on a slight angle < 3 degrees. This compensates for the drag of the wheels when turning and allows for much faster game play.

[AdamHeard]

Originally Posted by nparikh

We've been debating using a Nascar technique to robot design. After having realized that the most efficient setup for a 6 wheel robot is to lower the center wheel, we statistically analyzed all of our matches and noted that we turn left much more than we turn right, and as such we are considering making the wheels on the right hand side a tad bit thicker (perhaps a few mm), and having the left hand train on a slight angle < 3 degrees. This compensates for the drag of the wheels when turning and allows for much faster game play.

Wow, that's pretty in depth analysis. You guys are actually considering making the robot turn one way better than the other just based on driver preferancec?

[Jeremiah H]

Originally Posted by nparikh

We've been debating using a Nascar technique to robot design. After having realized that the most efficient setup for a 6 wheel robot is to lower the center wheel, we statistically analyzed all of our matches and noted that we turn left much more than we turn right, and as such we are considering making the wheels on the right hand side a tad bit thicker (perhaps a few mm), and having the left hand train on a slight angle < 3 degrees. This compensates for the drag of the wheels when turning and allows for much faster game play.

Nice job with the stats, I figure you will be using the same driver that you collected those stats on. Let us know how that works out, I like it... unless your driver/ game strategy changes...just a thought...

As for the drivetrain q, Tank ('06) was a six-wheel drivebase, with the middle being down 1/4''. We used 12''d, 4''w wet-track go-cart tires on the middle and back, and the fronts were narrow (to accomodate our pickup) 8'' pneumatics on our own beadlock-style rims. Our frame was welded, but we used bearing blocks to mount all our wheels, so warpage wasn't an issue. We located our CG slightly behind the middle wheel (partly by accident, I must admit), and this meant that we were generally running on our wide, tacky back wheels. We could push or at least hold everyone we came up against, but with the center wheel so much lower, it turned amazingly well. We didn't have any problems with wheel hop, for our CG wasn't above the top of our big wheels.
A lot has been said about experimenting with pneumatic tire pressure, but we didn't so much because we wanted as much tread as possible to be contacting the floor (running on a dusty floor is a GREAT way to check for this, by the way), but if you were using narrower tires it probably wouldn't be as big a deal.
I wouldn't recommend using four go-cart tires in a drivetrain unless you plan to be a "defensive specialist", as we were, or you can build a very light yet effective offensive apparatus (like our pick/unload roller).
I would recommend a six-wheel drive base if possible because you can get away from all the wheel hop problems of a four-wheel.
Good Luck, y'all. Let us know how it pans out. -JH
P.S. sorry, I tend to ramble.

[Scott Shaw 237]

In 2005 237 dropped their center wheel in a 6 wheel drive setup. They lowered the wheel 1/4 and it worked out perfectly. In 06 we also decided to go with a 6 wheel drive setup with the center wheel dropped. I can't remember 100% but I believe it was either 1/4 inch or 1/8 inch. Either way it worked really nice. The drive system was always beastly.

[Madison]

Originally Posted by Billfred

<offtopic>Can someone explain in businessmajorese how this would work effectively? It sounds interesting.</offtopic>

I've thought about this for a bit and, operating under the assumption that all the wheels remain on the ground all the time, I can't understand either how such an arrangement is helpful.

[Joe Johnson]

Originally Posted by M. Krass

I've thought about this for a bit and, operating under the assumption that all the wheels remain on the ground all the time, I can't understand either how such an arrangement is helpful.

I don't know if it is a joke or not. I have heard that back when he was a competitive bodybuilder, the current Gov of CA used to psych out his opponents by leaking crazy stories to the press about the "secret" to his success (e.g. eating huge amounts of some strange food or sleeping 20 hours each day rather than the hard work of actually bodybuilding).

Between this "larger wheels from front to back" posting and the NASCAR posting from Neil Parikh, it is quite possible that Team #25 is pulling our collective chains. But...

...this is how it may work and work quite well.

Put all axles in a line. Have the front/middle/rear wheels with radii X/X+a/X+a+b respectively where a and b are small positive numbers (both with respect to X and in absolute terms) and b<a.

This does provide a similar effect as lowering the middle wheel; on a flat carpet, only the front and middle wheels OR the middle and the rear wheels can touch the ground at any one time (I leave this as an exercise for the reader -- but trust me it is true). This changes the aspect ratio of the width/wheelbase and makes it easier to turn.

BUT... what else does it do?

It also means that the "ground speed" of each tire is slightly different (assuming that the wheels are driven by a single chain and the sprockets are the same size).

What difference does that make?

The first thing it will do is increase rolling resistance. When you are driving in a straight line, with two wheels on the carpet, it means that one wheel must be slipping on the carpet. This, in turn, means that the straight line rolling resistance will be higher than it would with the same size wheels. This sounds bad, but, when you have 3 motors per side like I understand Team #25 typically has, you have power to burn, especially when you are driving is a straight line.

BUT... ...what happens when this same robot turns? Now one wheel is already slipping on the carpet.

I remember a prof. in college beating one idea into us in his class, "Friction is a VECTOR and the MAGNITUDE of the vector is at MOST muXN." This sounds like a stupid and obvious idea, but it is surprisingly important. He would give us examples of where this is important like pulling a cork out of a wine bottle (by twisting the cork as you pull it out, you point some of the muXN Vector in a direction that is 90 deg. from the direction you are pulling... it is easy to pull out) and a rear wheel drive car on snow (if the driver guns the gas, slipping the rear wheels, very small forces toward the back of the car can easily move the car from side to side).

How does this apply to this case?

Well, it may mean that by having one set of wheels already slipping on the carpet, turning ability may be improved because there is a smaller component of friction available to resist the scrubbing of the tires on the carpet.

I am not sure that I have convinced myself that it is a good idea, it would take some noodling to figure out the right values of a & b for example, but I am not sure that it is as far fetched as I had initially thought.

Joe J.

[Wayne C.]

Gee Joe- Us Joke?

would we do that?

actually- everybody asks us how much we lower the center wheel and we always tell them that we dont lower it at all. But then we are told we NEED to so I guess our bots just dont work.

We don't lower the wheel. The robot overpowers lateral friction to make the turns. The wider the wheel base the better it works.

No matter- the Evil Machine base design is being revamped for 07 to be "a bit more powerful". There were a few bots this year that were able to move us when we had the brakes on so we want to have a little more "umph". We gave a bunch of teams the schematics through last season so we know there will be others like us out there this year. It is still a great basic drive design- just not strong enough for some applications (like pushing free of a corner goal).

WC

(PS- I have some kidnapped Moe Sticks sitting here in with a pile of FIRST mailboxes if you want them)