Kit Chassis + Mecanum = Flex

Some of the NERDS were helping team 3194 at the Arizona regional. One problem they had was with drive chains falling off. The robot is pretty simple, it uses a kit chassis, with AM mecanum wheels driven with the kit style toughbox and mounting design, one motor and transmission per wheel.

The design of the chassis, transmission mounts, and wheel towers allows some flexing. This is made worse when the towers are cut away for wheel clearance.

I helped the team analyze the problem, and this is the fix they came up with. Simple, effective.

Also, notice the sleeve in the picture, this is the best dressed team I’ve ever seen! (Thanks to their sponsor JC Penny)





Nice fix. I vote two bolt holes on the toughbox side of the brace to remove that last degree of freedom, but it’s likely overkill in this situation.

Flex + Mecanum = good :smiley:

But it looks like they had too much of a good thing. Putting that bar essentially along the chain path is a wonderful solution. It keeps the distance between sprockets from changing, which is exactly what they needed.

A nice fix… and we found the flex helped our mecanum keep all four wheels on the ground a few years back… but then again, we were using direct drive.

The extended output shafts from AM and Banebots make it easy to mount wheels directly on the gearbox (so long as the outboard end of the shaft is supported, at least) and, while it does reduce some of your options in terms of gearing and drive geometry, it ensures that you will always have at least one wheel turning on your robot!

Jason

The kit chassis flex isn’t limited to just mecanum. Using two 8" kit rubber treaded wheels in front, and two 8" kit plastic treaded wheels in back, our wheel supports flexed and bent enough for our chain to pop off 3 matches in a row.

Everything seemed to work fine for our first 5 matches, no issues with chain or driving. After that 5th match, and what we now suspect was a rather fast/hard landing while traversing the bump, we noticed our drive chains started losing tension. We tried compensating by raising our idler shafts to maintain proper tension, but during each of our next 3 matches, the chain still popped off within the first 30 seconds. We even tried removing a link, and tensioning to the point where it was almost too tight, but the chain still managed to fall off.

During this process, we noticed our front right side wheel axle wasn’t straight. Looking closer we saw the outside of the wheel support had bent towards the center of the frame by almost half an inch. We managed to straighten it as best we could, but in the next match the chain popped off again, and the wheel support was bent in the same direction, after carefully and slowly driving over the bump. We’re actively working on a method of reinforcing the wheel supports to keep this from happening again at the Connecticut Regional, and some form of c-channel between the axle and frame seems to be the quickest, lightest, and most durable solution. If we have enough time, we may try to design entirely new wheel supports, with a more reinforcing angle between the wheel axle and the center of the frame.

Most of the flexibility stems from the fact that all the parts are channels, there are no box sections. Consider putting a plate across the open side of the wheel support tower, firmly attached to both sides at several places. This might prevent the towers from being able to twist. When the wheel (and sprocket) are at an angle, the chain will want to pop off.

Also look into adding the tubular link between the transmission and axle, as shown in the picture.

We were at the same regional and were throwing chains like crazy (lost six points in on match dragging the chain around behind our bot). Wish we had talked with you. Is that brace made out of EMT? We may try this approach along with some home made chain guards at our home regional on the 25th.

This is sort of changing th subject, but my team had trouble getting chain to stay on this year, but it hasn’t been an issue before. Do the highly intelligent people on CD think that it is because the wheels were on the kit standoffs and those flexed enought to allow the chain to pop off?

thanks!

That would be my guess.

Multiple teams I know of had this problem.

Not enough information to make this assessment.

Yes, the pods could have contributed to the chains popping off, but thats not proof. Many teams (mine included, without tensioners) ran chains to the wheel pods without a chain ever popping off.

Can you give us more info about your chain routing/tensioning? maybe even a picture?

At SD and SVR, I saw many problems like this. I had tried to help out a team using the kit chassis. They couldn’t get their chain on because the standoffs had bent too much. And whenever they did get the chain connected, it came off during a match because the standoffs would always flex too much.

It was one of the reasons our team chose not to use something like the KOP standoffs to mount our four wheels.

We ran a 6wd with chan runnning from the gearbox to the back wheel (we got these to be tight enough and it might have fallen off once). From the back wheel to the middle wheel was the problem, even with the AM tensioner on it, it consistnetly fell off and we could never solve it). From the middle, the chain went to the front wheel, which was very tight and we never had a problem.

Unfortunatly, I do not have a good picture. The one I attached is the best I could find. The lower set of wheels shows the gearbox through the middle wheel. The upper set shows the front wheels, which were not a problem.





I saw setups like this on a number of teams that had chain issues (and actually saw 25 chain breaking).

From what I could gather the AM Traction (Plaction?) wheels, with the tread attached, has a slightly larger circumference than the other wheels. Since both have good traction (.9+ CoF) the difference in distance traveled causes the chain to jump over time (~ 1 minute into matches). Try putting slick wheels, or zipties (Lower CoF) on the front wheels (the set chained only to the middle)