center-wheel drive

hey there! i’m surprised so many people are on the forums so far from the FIRST season – why all the buzz?

anyway, i’m just wondering if anyone has ever used a center-wheel drive, with two drive wheels in the middle and casters on each corner like this:

o--------o

--------

()-------()

--------

o--------o

where the 'o’s are casters and the '()'s are drive wheels.

my team seems heck-bent on using center-wheel drive, so i’d like your input. does it work? are there any problems/quirks/limitations we should account for in our design? would you reccomend center-wheel at all?

thanks for your help!

~ aaron

way back in '97 we built a center wheel drive…i don’t recommend it.

in my opinion drivetrains have been a major deciding factor in every FIRST contest. You can develop new systems in the off season and repeatedly develop designs year after year, so many veteran teams bring a lot of drivetrain experience to each new game. With small bumps or ramps being the biggest obstacles to overcome most teams can adapt their highly advanced drivetrains to new challenges.

Center wheel would not be my first choice. It does offer turn-on-a-dime capability, but it offers an opportunity to get high centered easily. We experienced this in '97 near the lip of the spinning tree. This is potentially true with any two-wheel drive vehicle though.

Just my opinion!
Put four on the floor, or spend some time developing a more diverse and adaptable drivetrain.

this year’s combbat robot used this type of drive train.
while it performed better than most other two wheel drive robots (the average configuration of 2 powered wheels and 2 casters) when it was on the hill, i did see it get stranded once when it ran up onto another robot, which lifted its drive wheels off of the ground.

Any time you have non-powered wheels (i.e. casters) supporting any of the robot’s weight, you are not getting as much power to the carpet as you possibly can. Therefore, if the game requires pushing force to gain an advantage, I would recommend that any surface that touches the carpet should be powered. If power is not an issue, than it doesn’t matter much.

-Chris

Casters = a big no

Our first year we had a four wheel drive bot that was overweight by about 10 pounds. The only way to drop the weight was to drop two of the wheels. We figured we would be real smart if we simply replaced them with two casters. With the casters driving the robot became more of a chore. Instead of focusing on the game the driver had to focus on just getting the robot to drive straight.

A lesson learned on our team is that all points touching the ground will always be powered.

Aaron-

There are some great attributes to center drive designs. The two biggest imho are simplicity and maneuverability. If you want to be able to spin in place it is a great system because you have very little scrubbing. Because you only have 2 drive wheels it is simple and lightweight.

But… like the others have said, it won’t work if you are going up ramps or over obstacles. You get high-centered. If you are in a pushing/pulling contest you HAVE to make sure you get weight on the drive wheels and NOT on the casters. Casters will kill ya if you don’t watch out.

Every year our team has done center drive (1997, 1998, 2000), we have made it to the national semifinals, so it must be good…:)) BTW, in 2000 our center drive was “enhanced” with shift-on-the-fly rear-wheel-drive wheels to get over the ramp, so you can’t count that one as simple.

Ken

P.S. The reason we are on here so long before the season is we are, uh, geeks. And I mean that in the most complimentary way!

ah the infamous 2 wheel drive topic… my high school team, team 263, has used 2 wheel drives ever since 2001. In 2001 it was a 2 wheel front drive, 2002 was 2 wheel center drive, and 2003 was 2 wheel back drive. The difference we did was, no casters… we used teflon skids. In each place you would put a caster, we put a curved peice of teflon since that was the best material we found with low friction that didn’t get scratched up easily. The skid system won us a Delphi award in 2001 and a Motorola award in 2003, and our 2001 robot also won the long island regional in 2001. 2 wheel drives provide you with a much greater ease in turning, but thats pretty much the only advantage you can get from them. With the concept of having a caster or skid on one end or both ends, you will have weight resting on these skids or casters, thus casuing less weight to be on your wheels which ultimately leads to a lower max pushing power. Some recommendations I give if you are set on making a 2 wheel drive would be:

  1. keep as much weight as possible over the wheels, ie. battery, compressor, add iron under the axel to bring the robot up to max 130 pounds when done building, etc.
  2. find a low friction non-caster material for the skids, avoide casters at all costs. especially if we get a field element like the ramp mesh from 2003.
  3. add a wheel front and back of each center wheel to make it a 6 wheel robot :smiley:

In all unless you get a game like 2002 where you can grab 2 goals and balance your self out over the 2 wheels, I would not recommend a 2 wheel drive robot. However, it can lead to some very interesting design ideas and innovative awards, with the added bonus of easy driving.

Casters are the worst thing to use for a driving component of a robot. Use tank drive, it is more effective and very easy to build.

Basically, attach a motor to a chain that is connected to the wheels on the left side of your robot, and then connect an identical motor to a chain connected to the wheels on the right side of your robot. Very easy, I am sure there is a whitepaper somewhere with more detail =)

Casters are the worst thing to use for a driving component of a robot. Use tank drive, it is more effective and very easy to build.

It depends on what castors the robot uses. Ball transfers probably would make the best castors since all they are is giant bearings.

I will alert some one from Team 166 about this thread, they used center wheel drive last year with moderate success.

Do you want to make a sports car or a truck?

It all depends on the game. For 2002, there were no ramps, so there were a few very successful 2 drive wheel bots, 95 included. In that game, we decided speed and maneuverability was more important then traction. In addition to giving us great handling and letting us devote the whole front end to ball handling, the drive train was almost an after thought in terms of design and work time. It let us focus on the much bigger problem of the ball pickup.

I’ll second the ‘skids in place of casters’ idea. We used something along those lines, and it worked out very well. I would also suggest trying not only to put weight over the wheels, but to balance the weight as close to 50/50 (front/back) as possible. This way you put as little weight on the skids as possible when accelerating and slowing down.

It has it’s very real advantages, and its very real drawbacks. But, like I said, what kind of bot do you want to make? That depends on the game, which I’m convinced depends on the mood Dean’s hair is in the day they actually start working on the game.

So, if you don’t want to build the typical pushing 'bot, then two wheels are a real fun way to go. You’ll have the chance to move your focus away from drive train design and onto all the other stuff that sometimes gets ignored. If nothing else, you’ll have fun driving it. Two wheeling is just a pleasure to drive.

You might also consider a 6 wheel design, with the center two wheels set slightly lower then the outside 4. This gives you great climbing and traction with (hopefully) much lower scrubbing when turning then a traditional 4 wheeler. Its a good mix, but is a bit more complicated.

-Andy A.

wow! thanks for all the feedback – i can’t believe you all responded so quickly!

i noticed a lot of people are saying that four-wheel drive is the way to go. how do you do it? we tried four-wheel in FIRST last year, and it was a disaster! we couldn’t turn at all until we covered the rear wheels with slippery plastic to decrease traction, and even then turning nearly burned out the drill motors. in the end, the only way to get decent manuverability was to run each wheel off a seperate motor. what did we do wrong?

p.s. what does “high centered” mean? i’m afraid i’ve never heard that term before

again, thanks for all your help!

*Originally posted by Mercutio *
i noticed a lot of people are saying that four-wheel drive is the way to go. how do you do it? we tried four-wheel in FIRST last year, and it was a disaster! we couldn’t turn at all until we covered the rear wheels with slippery plastic to decrease traction, and even then turning nearly burned out the drill motors. in the end, the only way to get decent manuverability was to run each wheel off a seperate motor. what did we do wrong?

From the sound of things, I’d bet you had two problems: your gear ratio was too high, and the treads on your wheels were too grippy. Those two things together can cause big problems for any robot. However, if you pick the right tires and the right gear ratios, 4WD can be many times better than 2WD for pushing power, especially.

Here’s what you gotta do:

Take the 4 wheel drive and the center wheel drive systems and combine them into a 6 wheel drive system. Your turning problem will be solved by slightly lowering the center wheels. You’ll have no worries about castors messing you up, either.

Just power the center wheels directly and connect the outer wheels to the center ones with belt or chain.

This is the most effective of the simple drivetrains. (I consider holonomic, swerve, ball, and tank drive to be complex.)

George

in 2002 we went to championships and we had serious drive problems with our tank drive 4 powered wheels on two motors … at some point we decided to go with caster wheel dead center in the back of our robot causing two of our power wheels off the ground and leaving two on the ground to move around with… we were fast robot to beginning with … once we ran with the caster we went wild on the field we spun around about faced and out maneuvered the other teams in such great way tho if u do remember the 2002 game was a flat field game with only goals as the only real structures on the field so we had alot of running space and made good use of it

you might think that caster would make your robot go out of control but with the lifted wheels on each side of the caster every time we made turn the wheels feel to the ground and made contact with the ground breaking the turn and keeping the robot from going all over with no control so if done right it suggest the use of caster when “battling” and out maneuvering other robots is involved (it also makes great lil hat trick when the other teams trying to pin you to wall with super easy and quick about faces ) and when speed is needed over power cuz with caster u might end up with the short end of the stick everytime u push against others

Casters are great !!! just look at my thing below lol:yikes:

caster…no

every how and then I have some imput that is worth while. How I think ill share that.

In this years game we experienced many robots with casters and and center drive. We were thrilled when we got to go against them, and disapoointed when we got paired with them. Why? Caster-bots can turn very quickly and efficiently. And if the game was all about this then it may not be a bad idea. But what you must know is that If I hit your robot anywhere but the front, I will spin you and then push you. You are also going to lose alot of ability to push other teams as well. The more wheels on the ground, the better, this will give you the most resistance to opposing motion.

One team that had a very sweet caster-design was team #9. This team rarely gets the credit they deserved but their robot performed perfectly. They set 4 wheels in the back of their robot (all powered) and had 2 casters in front for balance. They were able to turn effortlessly and were able to resist pushing and pulling quite well.

If you want mobility and power you can always use crab or omni drive (which are a little complicated) but otherwise you could use what we did. We had four wheels on the ground and had a caster attached to a globe motor that could deploy in under a second. As a result we could turn quickly and then pull the caster up for power. We were amazing at defending the stack, and I attribute it to this design, and awesome driving.

hope that helped

if you would like to use the ability of castors, and get the advantage of being able to have crazy manuverability… try omni-wheels. my team experienced with these last year. what it is if anyone hasnt seen these before, they are a small wheel and they have rollers made into the sides of the wheel, thus acting as a castor. the advantage of these tho, is you can power them, and not be screwed out for manuverability.

*Originally posted by Mercutio *
**wow! thanks for all the feedback – i can’t believe you all responded so quickly!

i noticed a lot of people are saying that four-wheel drive is the way to go. how do you do it? we tried four-wheel in FIRST last year, and it was a disaster! we couldn’t turn at all until we covered the rear wheels with slippery plastic to decrease traction, and even then turning nearly burned out the drill motors. in the end, the only way to get decent manuverability was to run each wheel off a seperate motor. what did we do wrong?

p.s. what does “high centered” mean? i’m afraid i’ve never heard that term before

again, thanks for all your help! **

It is not uncommon for teams to try 4-wheel drive and not be able to turn. I’ve been working on a 4-wheel drive white paper that shows the calculations you need to make to ensure that your drivetrain will turn. I think it will be done this weekend, so I should be able to post it Monday morning.

“High centered” means that your robot is stuck on something (typically in the center of the robot). The term comes from 4x4 off-roading. Think of it this way: you’re driving your Jeep through the woods and you attempt to drive over a large log that is across the path. You get halfway over the log, and your Jeep is now resting with its frame on the log, and your wheels can’t get traction. This is because you have something high in the center of your vehicle - hence, your vehicle is “high-centered”.

-Chris

I was too lazy to read all the threads so I hope I’m not repeating something but it is very possible to be successful with a center-wheel design. In 1996, team 73 used a center-wheel drive. The center wheels were just powered wheel-chair wheels. There were custom-made spherical casters on the four corners of the robot. The chief advantage of the center-wheel drive at that time was manuverability. This meant that it was possible to spin out of an engagement with another robot. I should also add, that 1996 was the year that 73 was national champions.

Matt

the main problem with a center drive 2 wheel bot is that when going up a ramp, the casters in front will raise your drive wheels off the ground, getting you stuck. 2 drive wheels in front and casters in the back arent that bad though (omni wheels in the rear are even better)