pic: Annotated Chassis Drawing

This is an annotated drawing of our Fall project that helps potential sponsors to understand what the different parts we use are, how they work together and why we need their money. :slight_smile:

Looks impressive.

A question about the module pivots, though: is there anything inside the frame to reinforce the pivot points? It looks like you’ve got a 7/8" hole in 1"×1"×1/16" aluminum tube. Especially on the perimeter of the frame, that has the potential to get rather bent during competition, unless reinforced.

While I have your attention, is this intended to be used with a sprung suspension, or a powered mechanism to drive the pivots?

Looks mighty slick. Two questions-
-How’s traction/“pushing power” in the front/back and side/side directions?
-How’s maneuverability?
Keep us posted on how it works out- our team may be interested in that sort of drive this year.

Along with what Tristan said, the frame looks fairly light. I wouldn’t be surprised if the frame flexed enough to make do without the drive module pivots. Though, knowing your design skills, I’m assuming you’ll have more structure going on top to stiffen the whole assembly, hence, making my point worthless. :slight_smile:

Do you plan on rivetting/welding/bolting the frame?

However, I’m at a loss as to how the tensioning slots work. Since the idler moves with the gearbox module, it seems like all you’re doing is changing the tangent point of the chain with the wheel. I would expect the idler to move with respect to the gearbox, or to just use an eccentric idler/guide. That could greatly simplify the entire assembly.

Overall it looks very nice and clean. One of the better Mecanum setups I’ve seen.


Excellent CAD drawing.

If I was a potential sponsor, I would briefly like to see a work flow chart, concepts, designs showing different phases of development.
I would also like to know how the parts of the drive Tran can be fabricated, the work plus cost involved with the machining and assembly.
The possible use of outside services to perform any advanced machining of parts, the testing phases and the redesign phase that the teams must go through. Tie it all up with the learning and skills the students are receiving from working on this design.

This all translates to “what am I getting for my sponsorship dollars and how the money benefiting the students.”

(That’s just my opinion) Again, very nice design.

WOW that kinds looks like our new drive train except our new drive only has a total of 3 wheels. We figured we will need the weight this year for a lifting mech.

The pivot itself is a 3/8" Grade 8 bolt. The 7/8" hole you see if clearance for a ratchet, but it may turn out that we’ll run the bolt through both walls of each tube. In the past, we’ve had some overzealous kids overtighten bolts and crush tubing more than any robot has ever done.

The suspension is ‘sprung,’ as it were. The first iteration will use the compression of the rubber dampener as its spring. I’m hopeful this will allow us to easily change the resistance provided by the suspension as well as ride height by altering the properties of the dampener.

Thanks. :slight_smile: These sorts of drives aren’t built for fighting other robots. If you get into a place where your mecanum or omnidirectional drive needs to push another robot, you’ve probably made a bad decision somewhere, I think. We’ll have a better idea for what the traction and losses will be like once we have the wheels on hand – hopefully by the end of next week.

We built a very, very light frame for the 2006 game that consisted of four pieces of 1"x1"x1/16" Aluminum tube notched and bolted together to form a tic-tac-toe pattern, onto which we snapped drive pods that had omniwheels at the corners and Skyways in the middle for a 6WD arrangement. That chassis has been beaten up pretty badly and isn’t square or level anymore, so I’m anticipating similar abuse here and including the suspension. I don’t have anything planned for above what’s shown for our prototype unless it happens that we need to stiffen things up considerably more, but of course for competition there’ll be a lot of structure added.

We have to outsource welding to a local community college instructor (who sometimes likes to let his students practice on our parts) and so I’m hesitant to design anything at all for welds. The plan now is to bolt the frame together.

The tensioning slots are tangent to the wheel and sprocket, so by moving the gearbox up we change the center-center distance between the sprockets. The chain length can be changed by about 2 links by moving things up and down.


Does that clear the confusion up? The “idler” will be driving an optical encoder, so I want positive engagement with the chain and adjustability elsewhere.

We develop these plans each Fall as a way of giving students hands-on interaction with tools in our lab before the season begins. We have access to some pretty sophisticated tools, each with a unique learning curve, and so we’d waste a lot of time teaching their use during the build season. We’d make all of the parts ourself except, obviously, for the COTS stuff. We have a HAAS CNC machining center, a DynaCNC plasma cutter, a CO2 laser cutter and a big 'ol lathe in addition to more typical things like bandsaws, drill presses, and the sort.

We chose a mecanum chassis because of the close integration required between mechanical, software and electrical components. We’ve become pretty good at making sound mechanical devices, so it’s now time to learn more about control systems. I think this will be something that will engage a lot of the students who so far haven’t been interested in getting their hands dirty with the power tools. It’ll also end up as an impressive showpiece, we hope, for demonstrations and outreach.

Depending on how well this goes together, we’ve got some other designs coming down the line that we’ll take a stab at before the year ends. Those projects are a bit more traditional, though – an improved 6WD design, an 8-speed, shift-on-the-fly transmission… y’know, boring stuff :wink:

Are you using traditional omniwheels for a three-wheeled drive or mecanums in some arrangement that I can’t wrap my head around this time of the morning?

I sure hope we don’t need a lifting mechanism in 2007. We decided last night that we’re tired of driving around giant, cumbersome robots and our next machine will be two feet tall and seventy-five pounds, no matter its intended goal :stuck_out_tongue:

I don’t think this is as strong as some people think it is there is no cross support for the box tubes that make up the wheel wells, hit the side of that robot enough times it will start to bow in trust me.

I like your idea for suspension, that was a major problem on our robot last year. We had to go 2 wheel drive to save weight at the last minute, and at all of the matches our robot was out of square (despite our wheel adjustments). It was a real pain to drive.

It might just be the way I’m lookin at this, but it seems that all of the rollers on the mecanums are in the same orientation. Isn’t the concept of mecanums to have 45 degree offsets?

You’re right – I’ve just been too lazy to mirror the model of AndyMark mecanum wheels to get a complete set.

What code are you using for your setup? We’re working on a mecanum setup very similar except that we’re having problems programming it. Especially turning and strafing at the same time. Sorry if this is a little off topic

I was wandering if you made the AndyMark Wheels in inventor or down loaded them from some where? If you made them what did you do to get the angled tabs right. :confused:

I downloaded the model from the AndyMark website.

What code are you using for your setup? We’re working on a mecanum setup very similar except that we’re having problems programming it. Especially turning and strafing at the same time. Sorry if this is a little off topic

I’m sorry I didn’t see this sooner. I know you guys seem to have finished up your drive, so would you still like a look at our code? I think I have a copy of the completed code, but since I don’t do any programming at all, I can’t be sure until I harass our programming folks.

Currently we are using a modified version of some vex holomonic code that I found on chief delphi. It works but its hard to control and I’m sure there’s a more effecient way to do it.