pic: 114 Tank module design

What do you get when you mix three mechanical Geeks, some food and drink, a computer with Inventor 10, and Brecoflex and Martin Sprocket catalogue?

A tank tread attempt. I’ve been meshing my gearbox with different designes lately, and here is the most recent. Gearbox is the same as the last one (single speed, output speed of 6.7 fps), but this time it’s hooked up to something far more traction-y.

As always, comments/criticism/questions more than welcome.

6.7 fps is a little slow isn’t it? Also, why so much of the output shaft overhanging the side of the frame?

It’s a bit slow for racing, but the tourque allows good pushing power in exchange. The shaft is a previous iteration, and hasn’t been cut down to size in the model yet.

Wow… That looks so sturdy it isn’t even funny. And Very aesthetically pleasing as well. How much does it weigh?


we made that mistake this past year, im advising everybody… NEVER RELY ON FRICTION BELTS!!!..

We shouldve known better…

The weight, minus the motors, should come out to be 10-12 pounds. And yes, I did design it to be very strong. I want our drive to be unafraid when he rams away.

The belts we are planning on using are from brecoflex, with the ridge in the middle to keep the belt centered. The backing, or the part that will contact the ground, is red linatex, which has been rated on FIRST carpet to have a 1.6 coefficient of friction.


However, if I may, I’d like to offer some constructive criticism (really doesn’t matter for functionality) but why not turn your transmissions upside to maximize space about the tracks for your superstructure? In doing so it would put your COG closer to the ground helping your system stay upright. If there’s a constraint issue, by all means I applogize, i can’t tell from this angle.

Looks impressive, Kudos.

Turn the transmission upside down… I hadn’t thought of that. That would actually make things nice and compact, I think I’ll give it a try. All I’d have to do it switch one dimension, and flip the thing over. Thanks for the ideas!

[Edit] I flipped it, and it turns out the gearbox is too tall, and stick below the treads. ah well. [/edit]

Pretty cool you guys…

I think Andy’s speaking more about how the belt is driven – suggesting toothed timing belt over flat belt. Timing belt is what most teams use, though it’s possible to use flat belt if the drive is well designed. For the 2002 season, I helped design a very high traction tank drive system that used flat belts, for example.

My concern is that there’s no apparent mechanism to tension the belt. My understanding is that belts need to be ordered with the expectation that they’ll be stretched into or onto the drive. A robust tensioning device will allow you to more easily add or change belting and adjust the tension to prevent breaking the belts.

[Edit] I flipped it, and it turns out the gearbox is too tall, and stick below the treads. ah well. [/edit]

How far does it extend beyond the plane of the floor? It looks like you ought to be able to invert the gearbox and raise the height of the drive pulley from the floor to accomodate the new position of the gearbox.

why build a gear box at all?

Use the side-plates of the treads as the side-plates of the transmissions. Then you would have everything in a nice compact package.

Tensioners aren’t really necessary if you use the correct spacing and steel reinforced belts. The other track-drive system currently posted on CD media, is a tried-and-true drivetrain that uses no tensioners.

By the way, those look like some pretty tight turns. We have a spare piece of berco-flex belting, and just from twisting it I doubt it will be able to bend that much.

  1. Why is the wheelbase so narrow? Even without any weight on the top, I can see this drivetrain constantly rocking back and forth onto the steep incline/decline sections.

  2. Do the two, lower idler pulleys extend beyond the lower edge of the bottom, flat plate? If not, you’ll have some excess friction on that plate, which will kill any efficiency you have. You’ll need to lower the idlers, or put more idlers in the center of that plate.

  3. Ah, now I see the tight wrap around the driven pulley and it’s two idlers. I guess you’ll have to read the specs on your specific belt.

  4. Using a double motor gearbox, you should be able to gear that a little higher and still have the same pushing force. I would imagine that you’d still slip the belt on the carpet with a bit less torque…

Looks good!


I also am a bit concerned about how small your wheel base is also, on of the greatest advantages of a tank system is having a HUGE wheelbase. I can see that the reason you made it so small is to be able to climb something… but you wont need that great of an angle to climb anything. Otherwise it is a beautiful system which I would love to see in person if completed, but I also recommend using timing belts… The transfer of power will be much more efficient and you will not have to worry about slipping (thou I doubt that would happen).

Thanks for the feedback!

  1. The wheelbase was so narrow to allow the bot to drive either direction over an obstacle, and not have to turn around in order to climb steps. This has been fixed.
  2. Now they do.
    3.This has also been changed to give less tightness to the belt. (my words don’t make sense to me today…)
  3. We currently like this gearing, as price, size, and machining capabilities allow us to do all the work in house, which saves on time.

For those that care, I added another iteration.