pic: 1097 Direct Drive 01

b705e0bfc6e9bae212e7d32e36577243_l.jpg

They said it couldn't be done effectively in six weeks. They said it would break under the torque required to drag the outer wheels around. They said it would look pretty but be ineffective. They were wrong.

Our drive train this year is a six-wheel direct drive transmission. It is powered by two CIMs on each side (you can only see one in this picture) mounted across from each other. The center wheel is lowered 1/32" which has proven to be very effective in lowering torque while turning without too much rocking. The wheels are custom CNC with the treat recessed to make them more like rims. The U-joints are to let the shaft be non-linear to accommodate for the 1/32" drop in the center.

I love my chief mechanic.

woah,was it worth having no chains?

whats the weight comparison to chains and sprockets? also does that have any advantages like less moving parts or more giving? interesting

  1. Very cool. I’m impressed.

  2. What you have is a shaft-drive system. “Direct drive” in most places I’ve seen it would have the power module (in your case an electric motor) directly driving the wheels without gear (or belts with pulleys) reduction. I do not believe “direct drive” is the best description for your bot.

  3. Did I say I thought it was cool? I like the way it simplifies the packaging of the drivetrain.

Ouch says the inefficiency meter. Have you guys calculated the losses in all those 90 degree reductions? How are things set up inside that gearbox, is it all straight cut spur gears? Even still, it is really cool and unlike anything I have seen on a FIRST robot yet. +1 creativity points.

Ooo. I like. Very unique. Are those driveshafts aluminum? Also, what is the efficiency on mitre gears?

It looks really pretty, but as SpaceOsc asked, why?

Sanddrag, bevel gears (I believe that is the more common name) have similar efficiency to spur gears (95%+) because they do not have any surfaces sliding across each other (like in a worm gear).

My reason would be there is no possibility of throwing chains or broken chains. Also, my favorite reason for it is no chains to tension. And there’s also the “it’s really cool and hasn’t been done before” aspect.

Well, to answer the efficiency question, it’s very efficient. The miter gears are aligned within .0005" and transfer the power better than chains. We have had chain problems before. Last year we threw Motorcycle (40 gauge) chain many times, so it removes the need to tension or worry about chains. We went through the entire sac regional without damaging it at all. The ony thing we had to do was check the set-screws to be safe.

As for the drive-shafts, they are .049" wall steel tubing.

As for the efficiency on the miter gears, they are just as effieient as straight-cut spur gears.

We lightened all the gears in the gearbox with 6 holes in each, similar to the wheels if you have seen them. Overall, the entire robot weighs 106.8 lbs, so we have no weight concerns.

And, as Sanddrag said, we did it mainly since no one else has done it.

My mechanic told me to tell you that there are 46 bearings and 42 set-screws in the drive-train.

And as for shaft-drive/direct-drive, we’ve always seen shaft-drive as a sub-set of direct-drive.

-Tony K

How is rotational load transfered to the U-joints? If the driveshaft was solid, I would put a keyway in it but since it is a tube I’m wondering how you did it. Did you put a pin through? .049 is pretty thin, thats awesome.

i saw it in your pits, very cool and unique. Congrats on doing something unusual!. And is that the kit gearbox?

The OD of the U-Joints were turned down to within .005" of the ID of the steel shaft. We then put set screws all the way through and have had no problems with them. Eventhough, we are still going to replace all 42 set-screws on thursday at SVR.

As for the gearbox, it is about the same ratio as the kit gearbox, but we couldn’t use the kit gearbox due to the way we had to lay it out. As you can see, we needed the driveshaft to go straight through the gearbox and the kit version wouldn’t allow for this. We therefore also had to put the CIMs across from eachother. This is useful in that neither side will be more powerful since in either direction, you have one motor running forward and one running backward.

-Tony K

where did you guys get the U join?

Oh okay. I was envisioning the tube going into the U joint, not the U joint going into the tube. Very clever! I hope it continues to work out well for you and maybe we’ll give it a try next year!

If you want to give it a shot, I’ll see if I can get my mechanic to give me all of the blueprints and I’ll scan them for you.

I am drooling. I hate chains. Just today I had to replace chains four times… That may have had something to do with the fact that I was riding on the robot, and the wheel mounts really shouldnt even hold 120 pounds, not to mention nearly 300 :smiley: .

So, I know it has been quite some time since this discussion was last updated but I thought I would note the followiing:

  1. This drivetrain has been absolutely flawless, both in competition and in post-season stuff.

  2. It took tremendous time and effort to make it work, but in the long run it was well worth it.

  3. I strongly encourage other teams to use this system in the future. (If you ever have any questions, just e-mail me or Kesich).

~ Christopher