So i’m doing some preemptive research for next years build season.
And one thing we seriously want to do is use mecanum wheels on the robot
whats some good feedback you can give me about them? are there any cons to them?
also, where is the best place to buy quality mecanum wheels?
There are two places to get mecanum wheels small enough for FIRST use. 1: Make your own. 2: Buy from AndyMark.
Con #1: they need a lot of overhead weight (4 wheels, 4 trannies, 4 motors).
Con #2: You have to get the CG right or handling will suffer. Suspensions may be useful.
Pro #1: Lots of maneuverability.
Pro #2: You don’t need a special frame, so if they don’t work, just switch to a 4WD or 6WD. You can even leave the trannies in place.
Pro #3: Cool factor.
Other things–you don’t get as much of a “push” factor. You need to avoid shoving matches.
OK, random thought: If you move the center of rotation based on where the (expected) center of gravity is, would that help the bot at all?
Nope. If the CG isn’t near (at) the center, the robot won’t translate straight. That’s what I was referring to when I said that handling would suffer. Center of rotation wouldn’t necessarily be affected for a turn.
There are a number of Minnesota teams that either used mecanum wheels on their robots either this year or last year. You could probably come down to the MN Regional and take a look and have a chat.
The two that I am positive have used mecanums either this year or last are 2052 KnightKrawler and (last year) and 2207 Prime 329 (this year).
I think I failed to express myself clearly. By center of rotation, I meant that point on the bot where it thinks its center is, the point that a and b are based on in the equation for one wheel’s speed, “|vw4 | = vty + vtx + ω(a + b)” (from strafing.pdf, page 19/35). I’m guessing that the controls would still come out incorrect, because of the uneven pressure on the wheels, but I thought I should make sure we were speaking the same english.
I would seriously discourage you from deciding on a drive train for next year before you know what next year’s game is. Mechanum has its place in some games, but not in others.
Some cons to mechanum:
Suspension is essentially required
High Weight
Low pushing power (fine for this game not for others)
Expensive/Time Consuming (you will either have to buy/make four transmission/wheels our mechanum promo-bot cost well over $1000 dollars for drive train alone.)
I think Andymark.com has good mecanum wheels. McMasters might also.
We aren’t huge fans for the mecanum wheels because of how much battery and weight they take up. Yes these are high speed high maneuverability wheels, but you need to keep in mind if you will have the weight to spare for a fairly heavy drive train. We find that the IFI traction wheels work just fine, light weight, plenty of grip for shoving matches or just pushing people out of our way this year. Yes turning isn’t as fast and they aren’t as cool, but we find they just work.
The advice to wait and see what the game is before you decide on a drive train is very sound. We did mecanum drive this year, and it drives well, but I would not have enjoyed trying to get this mecanum drive up some of the slick, narrow ramps last year.
That said, there are ways to cut down on the weight and cost. We made our gear boxes this year out of 1.25" x 4" aluminum tubing. Each box has two gears and a sprocket for the back wheel and two gears and a sprocket for the front wheel. This meant only two gear boxes and less weight. This year’s drive train components were no more heavy than last year’s 4WD. And the total cost was $375 for the wheels from AndyMark and about $220 for the gears, sprockets, shaft and tubing from McMaster. Not cheap but not prohibitively expensive either.
The most difficult thing we had to do was to balance the weight. When we first started driving, before the “arm” was done, we had to clamp a bunch of C-clamps to the frame to make it strafe straight. But we found that some prior planning made it fairly easy to distribute the weight properly. We have been satisfied with its performance in practice, but we’ll see how it does in competition this weekend at Buckeye. Hopefully it will work well.
Side note: Since this was our first year using mecanum, we also developed a more traditional drive train as a back-up. We ended up not needing it, but it helped knowing there was a back-up system. This development also gave some of the new students valuable experience and a chance for some of our younger veterans to get some leadership experience.
From a team that has done mecanum the past two years, Mecanum is an amazing drive.
First off many people think mecanum drives get pushed around alot or have no pushing abilities at all. This is completely wrong, we have yet to have any problems of being pushed around. This year (many of times) we ended up PUSHING ROBOTS!
As for the balance thing, This year we decied to use chain and sprockets rather then running direct from the gearboxes as we did last year. This allowed us to move the 4 motors and gear boxes forward or backwards acording to needing changes in CG.
Finally I’d say if your going to use Mecanum wheels us the AndyMark 8" wheels. Though the 6" are “smoother” and “cheaper” and “lighter,” they are weaker and more problematic. We used the 6" wheels this year and we had may problems. The way they where designed gives them pertruding edges that dig into the carpet. This caused us not to be able to turn in place. After purchesing another set of wheels and using 6 hours to remove and smooth off al the edges they worked great. When we got to our second regional and swaped out our old wheels for the new ones we realized that the hub area was now concave. The sheet metal is just too thin and the wheels way to problematick many of times we lost rollers this year, remidied by locktight.
Overall if you use the corect wheels and get the CG and code right your golden.
And yes Mecanum is bad for inclines
You can stabilize mechanum drive with a gyro without a whole lot of trouble, but, as previously stated, handling still suffers even though your robot won’t drift when strafing, since you’re using more motor power to make the robot not drift.
Also, the year before I joined our team (2005) our tetra robot had mechanum wheels but the software was never written to make it work like it should. So, that year the left two motors were paired to the left joystick, the right two to the right. Our driver that year said it was the best robot to drive he’d ever had, since it could powerslide really really well.
Just something to consider.
From personal driving experience with our robot interacting with other mechanum drive robots, i’ve found them very easy to push out of the way and if another is trying to push/disturb our robot, it’s almost not noticeable after the initial impact. Even in second gear our robot this year has been able to (so far) push most mechanum robots without much issue.
-q
We used mecanums this year and as a driver, I love them to death. They are perfect for this years game because if I get in a tight situation where I could possibly cross back over a line, I just strafe out of the way, which kept us close to no penalties at Kansas City.
We concentrated on a light robot this year, and we still managed to keep the robot under 80 pounds with bumpers, a battery, and individual suspensions. We did this by only using a 6:1 gear reduction, an 11 tooth sprocket on the motor and a 66 tooth on the mecanum wheel. This greatly reduced weight.
With the suspensions, it can be a good thing to have them, but they are not mandatory. The goal is to keep all 4 wheels on the ground so that the robot doesn’t start to go off to one side. They helped this year with the raised feet in the middle of the field. But I did acidentally flip our robot when i hit one of those feet wrong but thats neither here nor there.
The suspensions can cause a problem when strafing. If the CG is very high at all, the robot will tip slightly away from the direction you are strafing, throwing the CG off, but to fix this, just use stiffer springs to tighten up the suspension.
I wouldn’t decide on mecanum wheels quite yet, like Matt H. said. But it doesn’t hurt doing research now like you are.
I’m with you on most things here, Matt, but I’ll disagree on suspension. A “flexible” frame can do just as well.
My team looked at mecanum wheels from places other than AndyMark. According to the lead researcher, AM is the ONLY place that sells them in FRC sizes. So the only other option is make your own… That can take a while.
Oh, and Rupnick–I wouldn’t tell Flaming Chickens that. (At least, I think it was 1540 that had mecanums and climbed right up a nice plastic Beach’Bot ramp at SD 2007. If not, I think it was 1425.) You just said that they had a hard time doing what they did with ease.
Despite the advantages and disadvantages of any drive, I’d suggest waiting until you know your challenge before selecting any aspect of your robot design, with the possible exception of electronics boards (although for reasons best discussed elsewhere, I’d warn you against even that this upcoming year).
Also, keep in mind that Mecanum drives are not the only drive solution that allows for strafing. Both swerve drives and holonomic (“omni”) drives can achieve it, and have different advantages and disadvantages. I’d suggest searching around the forums for different threads and whitepapers dealing with each and make your decision based on the information gathered during that search and what the game demands.
Here are some useful starting points:
http://www.chiefdelphi.com/media/papers/1836
We used mecanum wheels that were very succesful… There are a couple of flaws.
They have great maneuverability.!!!though
Note that it is traction that COG effects more than anything. If COG is closer to a particular wheel, that wheel will get more traction and therefore more pushing force against the carpet. Since mecanum design is all about vectors that cancel each other out, it is imperative the weight be distributed evenly. Simple suspensions may help even things out, but movable weights and/or adjustable motors are probably simpler, lighter, and less exepensive.
After seeing a new concept of strafing that 1565 has put together this year, I am now not so impressed with mecanum wheels. They have great advantages such as simplicity to mount. However they do have their inherent flaws, such as a certain % of your power is lost due to power vectors that cancel each other out. This particular one is a doozy for me, and the new 1565 modification I have in mind fixes this and even omni drive’s difficulties in climbing ramps.
If you’d like to prototype mecanum before next year, the summer is the time to start. I’d take June & July to do research on parts, prices, and weight, then take Inventor or Solidworks and work out a frame & mounts. By the time the school opens in the Fall, see if your team has the resources to do a prototype (note that the parts & code you build now may or may not be legal to use in 2009) then I would suggest building the prototype. I would also note that the FRC controller is changing in 2009 unless something changes in FIRST’s plans, therefore any code you write now will have to be modified in the build season anyways.
If you require that your team use mecanum next year, you might be in a world of hurt when you go to integrate all of the pieces onto your bot. It is far easier to design things to integrate together from the start than to have an unchangable piece that every other piece must fit around. Frame modifications come to mind as well as manipulator mounts. That’s just my .02 for that subject though.
I’ve had some experience with Mecanum drivetrains, I’ll be inspecting at the MN regional, feel free to stop by and talk to me (Friday or Saturday, after everyone is done inspecting).
As far as ramps go, the only problem we had last year with ramps was not driving sideways off them if there were no sides. We solved that with a switch on the joystick that disabled the strafing, so that the driver didn’t have to keep the joystick perfectly straight while going up the ramp. The ability to line up on the ramp was priceless, then just pull the trigger and drive straight up. With a wheel diameter in the range that is useable on a FRC bot, a suspension is almost mandatory. Last year I was able to drive the robot over 3/4" thick obstacles on the floor at full speed when practicing.
Most of the pros and cons have been discussed pretty well in this thread. My advice if you are serious about considering a drivetrain of this sort is to get some help from people who have done it, and to prototype something over the summer to experiment with. Your programmers will appreciate that… We were able to make some plastic wheels on a rapid prototype machine a couple of years ago that we put on a Vex chassis. That let us decide that we could make it work on a full size FRC robot.
And always be prepared to change your mind and plans when you find out the game and rules for next year.