Mecanum Wheels is it good fin the Pits

Dear Friends!
We studded AndyMark Mecanum Wheels it is perfect for us, which can move in any direction, rollers attached to its circumference, these rollers having an axis of rotation at 45° .
But still we have suspicion if any robot will push us what behavior will be.
Second question what size wheels Left 8" Mecanum Wheel or 6" Mecanum Wheel for stabilization is good.
Any help will love to hear
Mohamed
Team 1946

As part of team 1083 we did omni wheels. It was awesome to work with and fun. Loved to watch the bot move around.

BUT you loose major pushing power maybe 45% where a few percentage points of extra push makes the difference.

With 1902 we ended up going to the basic six wheel design that was a strong pusher and turned well plus a lot easier to work with.

There maybe games where mechanism will be an advantage, but pushing seems to be a major issue most years.

You’re going to get pushed. On the other hand, you don’t have to worry about frying motors due to pushing matches–you’ll be spinning your wheels.

Don’t try to win any pushing matches; use the extra maneuverability to avoid them.

6" wheels inherently give you more torque over 8" wheels at the same motor power. They also have their mass closer to the axle making them (slightly) easier to spin. In other words, with the 6" wheels, you should see much greater acceleration but a little less maximum speed.

you’ll have awsome manuverability, get pushed around alot and very easily, and if you have to climb anything, like last years ramps, you’ll have some trouble…

Well… I’d hope they use the gear ratio to determine torque/speed rather than the wheels. but you are right…

As others have already said, with omni’s and mecanum designs, you will get pushed around. BUT what you will have is manueverability AND less stressed components. Since we have started using omnis on the outer wheels on our robots, our motors don’t heat, breakers don’t trip, less chain breaking/problems. But again we had to remind our driver all to often, not to waste time trying to push others around.

If you want to push folks around, your drive system better be tough.

Perhaps I should have stated “all else equal”, does that satisfy it? It was a small statement anyways. In the end, if we say torque/speed needs to be determined and set via the gear ratio then there is no reason to even choose between 2 different size wheels other than possible field components; so why even argue this at all?

The power transmission from motor to rotation against the floor is one great system that should hardly be considered internally modular for the purposes of determining the benefits of acceleration vs speed. As an example, wheels that have most of their mass away from the center can adversely affect either variable just as much as a bad gear ratio. It all has to be considered.

You will be pushed around since your mu is reduced by ~30%. The advantage is manuverability, and your strategy must require manuverability in order to make use of this advantage.

Second question what size wheels Left 8" Mecanum Wheel or 6" Mecanum
wheel for stabilization is good.

Either is good, depenfing on where you want your axles. Stability is much more a function of center of gravity than wheel size.

Any help will love to hear
Mohamed
Team 1946

Good luck and have fun

We had mecanums on our robot last year, they were great, we were able to play defense (between the other team and the rack so that we didn’t have to push), and if we got pushed, we couldn’t really push back, but we could easily dodge to the side and go around.

Try stacking 4 on a axle, it will give you more traction. Our design used a set of three at each wheel, it did help, we have not tryed four. We did put a student on a cart and we pushed him all over. I do realise that is different than a powered robot.

I don’t think that’ll fit… four mecanums would be like eight+ inches thick… I’m surprised you could even fit three on one axle.

As a part of team 357, I have been working with my team on different variations of Mecanum wheels for the past three seasons. One misconception we have found is that mecanum wheels cannot push or will be pushed very easily. However, we have found that mecanum wheels are good for what they are designed for. For example in 2005 we wanted a fast, highly maneuverable robot. We decided to use mecanum wheel with a fast drive train. Our design worked very well for us ad we were very happy with the results. In 2006, we modified our design to accompany a ten inch wheel, up-sized from our six inch wheel in 2005. By enlarging the wheel we got a much larger contact patch with the ground. As a result we found that we could push many robots, in all directions. In one competition we were able to push one robot, then another at the same time, preventing both from scoring in the upper goal. While mecanum wheels are not usually designed for pushing and are well known for their maneuverability, if they are designed right, and for that purpose they can be pushing force to be reckoned with, as well has having the known maneuverability of traditional mecanum wheels. If you want more more information our team is always glad help at www.team357.org or you can PM me.

Contact patch has nothing to do with pushing force.

Coefficient of friction (CoF or Mu) is what gives you your pushing force and maintaining contact with the surface allows you to sustain that pushing force in a shoving match.

More than likely your wheels were made of a different material (better CoF on carpet), you had a better suspension (able to maintain contact with the carpet) or just had more power (more motors or lower losses in the transmission). Because of the accuracy needed to shoot into the upper goal, in 2006, just a small ram (momentum only) was needed to ensure difficulty in shooting.

Mechanums, by nature, do not have the pushing power because the bearings allow for translation from an outside source (IE low CoF when being pushed).

For the past week or so I’ve been working on a design for a drive system implementing both mecanum and omni wheels

but right now im on the fence as to which way to go,

whether to allow for a complete omni system (ie separate motors for each mecanum wheel),
or to use two motors (one for each side) and then implement a transmission i designed which allows for the two wheels linked to one motor to change direction and stop (ie for the strafing and diagonal movements).

the down side to the second system is i will not be able to code it to drive in relation to the driver completely, but i believe it will still be possible to code it so it drives in reference to a side, (ie if the right side is facing away from the robot, it now becomes the front),
also the fact that if the robot is strafing or going diagonal, that that is the only movement it can do(ie to rotate the robot the driver would have to stop strafing)

the nice thing about the second system is you would have only two motors,
also the transmission seems to work well with the omni wheel set up, because i was able to align the gears so the only time the omni wheels run is when both mecanums on one side are going in the same direction, the omni wheels allow for the robot to climb which seems to be a significant issue with the mecanum wheels

the drive system would look like this
o/m o o/m
O-----O-----O

the first wheels being both an omni and a mecanum
the second being just and omni
and the third being an omni and a mecanum

all of the wheels on each side would be powered by the same motor, and therefore would go at the same speed

just looking for some feedback
thanks

Once kickoff is over, I will be posting up a new breed of omni drive systems (not mecanum though) which helps them with the pushing problem, if anyone is interested be sure to look out for it, I am putting the finishing touches on the design as we speak!

If you use mecanums on your robot, all you must is learn to drive with (not against) the force of other robots, this concept is applied in a lot of thing (martial arts mainly, as thats what I do for a living) and it will help you when driving.

Mecanum and omni are two very different drive styles than 4 wheel, 6 wheel etc… So if you do make it, do your best not to drive like a 4 wheel or you will get shoved everywhere!

One of your ideas is to use only 2 motors to each side. I question the wisdom of this, as you’re going to have to use much more real estate to get any kind of differential system in place. You might even reach the same amount of weight with a 2-motor system as a 4-motor system.
In 05, when we did mecanum, there was a small gear reduction to a chain reduction from the motor to the wheel, and it took up very little space. We were able to use field-oriented controls (with the old 150 degree/sec gyros and less powerful controller, so it should be pretty easy now) without pushing the weight limit.
Your design uses omnis to resolve the problem that mecanum wheels have with inclines, which is overstated. Last year, 3 or 4 teams with mecanum wheels were able to get up our ramps because they could drive straight. The problem mecanum robots have with inclines is if you try using the wheels as anything but straight on, you’ll have control problems when the wheels lose contact with flat ground.
In 06, team 40 used 8" mecanum wheels and were able to get up that 30 degree incline. They just needed a running start, and the fact that they had their chassis high enough to take the crest.

I would go with the one that will be rock solid so you will have fewer repairs and is easier to build ( cut down on assembly time) A drive motor for each corner is simple and safe. On our web site http://team1322.org/omni_drive.htm we have some instructions on how to build a simple mechanical joystick tie. The reason is as follows, in the coding for the robot drives there is a mix control for one joystick to tank drive. Use this to have the left joystick to control the left two motors. Copy and past it to use with the right joystick to control the right motors. With this mixing it allows you to go sideway when you move both joystick sideways, and allows you to go forward and back when both joysticks are move forward and back together. Now you can drive it like a two joystick tank control. The reason for the mechanical tie is to not allow the joysticks to move toward each other or apart. This is very easy to drive all directions with out allot of programming. We have tried all types of joystick configurations and this worked the best. I hope this was not confusing; I tried to make it a clear as I can with out pictures and diagrams.

yesterday I stopped by your shop to take a look at the robot (im a junior @ wpi), pretty cool stuff. I was wondering if you are the one who is currently working on the code for that system(your coach mentioned someone was revamping it) because ive never used the gyros or encoders and i would like to get a feel for how to use them well

just a note on the omni wheels, the idea behind those were that the robot would not need a head start, it would be able to climb up the 30degree rather easily, i remember seeing robots just tossing themeslves up there, and its a pretty scary sight. especially for robots like ours, that are always at the max height
also the nice thing about the omni wheels is, if we have an issue with the mecanums, we can ditch them in mid competition and just switch to a tank drive
i also thought it may help with issues with pushing, i understand that the strength of the mecanums is getting away rather than fighting, but if you could get the best of both worlds it would be pretty nice

thanks again for the feedback