pic: 2x1 gearbox



There hasn’t been a gearbox in a while so I thought I’ll post something. The whole frame of the gearbox is one piece of 2x1x1/16 box tubing with lots of holes. Adjusted theoretical speed is 12.95 ft/s with 4" wheels. This can be a separate gearbox, or it could be the drivetrain rail itself.

Chak is reminding me that there are gearbox designs I’m supposed to finish and haven’t.

As for the packaging, it looks great! I assume that the gear is exposed more on the bottom of the rail than the top because of the center-dropped axle, so would it be worth making this gearbox direct drive the back wheel instead of the center? That way you can flip the gear cut-outs and have the top side be more exposed. I honestly have no idea if that’s worth it; I remember we had issues with the large driven gears eating into the carpet on the 2015 Vex chassis, but that was with gears almost as large as the 4" omnis we had at the time.

After swearing to myself that I would never build another gearbox after designing like 30 of them, I’m finding myself breaking that pledge :stuck_out_tongue:

Might be my tendency to have reasonable doubt when it comes to structural integrity, but my concern is if 1/16" thick boxtubing will be strong enough, especially to support the CIMs.

This is a good start, when I was designing gearboxes one thing I would make sure to account for is how it would be mounted. I see two empty holes on either side of the output shaft, are these going to function as your mounting points for a face to face mount? If they do you will have trouble getting a nut or screw into the tubing because you will have a gear behind it. A good alternative might be to extend the ends of the tubing a bit and put mounting holes on the outside edge of the CIM motor. My last thought for this IF you are planning on a face mount is to use smaller screws and place them on the inside of the tubing. This will allow you to have a flat face mount without having to drill clearance holes in the chassis, except for the bearing.

I am becoming a fan of tube stock transmissions. Our robot included two of them this year, one for shooting and one for intake. Both used 775pro motors.

Thanks for posting this example using CIMs. We may try something similar, perhaps using 3x1. I also agree with those who favor 1/8" wall. Thinner walls make me worry about stress at the bearing seats and motor faces.

Looks very simple and that’s good. Actually it looks very Cimple, as in similar to the Andymark box.

Do you have any sources you’d recommend for designing gearboxes? I’m looking for information on design theory for the structure and so on. And please no gear ratio basics from the peanut gallery -_-

1/16" wall tubing is not a great idea for a gearbox like this, particularly because the bearings are only supported by a thin amount of material and also because you have to make an aggressive pocket to leave clearance for the gear. I’ve never seen bearings directly mounted in 1/16" tubing that support loads as large as a drive wheel work out well.

You really want to use 12T (or 11T) pinions on an enclosed gearbox like this because you are giving up the ability to use the CIM boss to pilot the gearbox when using a 14T gear, since your hole has to be big enough to clear the gear

FRC Designs has some examples (https://frcdesigns.com/designs-gearbox/ ), but for theory, I would start by searching CD for old white papers.

If the wheel is directly driven, the wheel would be concentric to the gear and it doesn’t really matter which way the gear faces. Currently, the 64t gear has a outer diameter of 3.3", so a 4" wheel should be fine. I actually wanted to use a 3" wheel for 9.71 ft/s, but later realized it wouldn’t work.
In the 3x1 version, the 66t gear has a outer diameter of 3.4", which should be fine also.

Done. :slight_smile: http://imgur.com/a/RNfht

Btw, both versions of this gearbox has a SRX mag encoder on the back. Using magnets means that I can loosen a shaft collar and take the output shaft out easily without dealing with the encoder. That makes me a little more willing to have holes blocked by the big gear. Still, as MattC9 said, it’s better to have mounting holes that are not blocked at all, so I added it for the 3x1 version.

At this point, this is just a WCP Single Speed Single Reduction gearbox with one less motors and less parts. :stuck_out_tongue:

How close are those top CIM mounting screws to the upper wall of the tubing? Are you having to remove wall material to thread those screws in? Would you if you increased to 1/8" wall tubing?

What’s the clearance between those same CIM mounting screws and the cluster gear?

What retains the output shaft from shifting?

It appears to me the hole on the CIM side of the tubing is sized to pilot the boss, and the 14T pinions are to be installed via the hole on the opposite side of the tubing once the CIMs are already screwed in.

There’s 2 sets of answers, for the 2x1 gearbox and the newer 3x1 gearbox. For the 3x1 gearbox, the clearances are huge (more than the diameter of the screw) and there is definitely no problem. For the 2x1 gearbox, it’s a lot tighter. The CIM mounting screws are .01" from the wall, and in theory no wall material will be removed. The CIM mounting screws are .0125" from the gear. The tight spaces are another reason to use 3x1.
If I use 2x1x1/8, the CIM mounting holes and the output bearing hole will cut the wall there down to 1/16 anyways, so I figured there wasn’t much of a difference. Using a 3x1 is much safer all around anyways.

It appears to me the hole on the CIM side of the tubing is sized to pilot the boss, and the 14T pinions are to be installed via the hole on the opposite side of the tubing once the CIMs are already screwed in.

Correct, except that I am using 11t pinions, so there is no problem anyhow.

What retains the output shaft from shifting?

The output shaft is turned down to a circle on the back end to prevent it from moving backwards. The circle extends past the bearing and a shaft collar holds it on the other side. It’s shown in the render in the link in my previous post. This setup allows me to remove the output shaft by removing only a shaft collar.

Thomas,

You’ve got a nice, simple design here. One more tip to add to the comments already provided in the thread:

You will need more space machined in the top or the bottom of the tube to allow for inserting the 64 gear into the tube. Unless you have some magical conjuring skills, that gear can’t currently be inserted into the pocket you have designed.

Sincerely,
Andy B.

It might fit in through a larger pocket on the bottom on the tube?

Just an assumption though- this definetly applies if there is an identical pocket on the bottom as there is on the top.

I haven’t designed or built a custom gearbox myself, but a few comments based on basic physics:

I like the simplicity, though I’m a bit leery about a gear ratio of less than 6:1, especially if your robots tend to barely make weight. I see this as much more likely built in to the chassis rail, especially as I try to figure out how to mount this segment of tubing to a separate chassis.

I agree with KohKohPuffs (and others) that 1/16" seems rather light for most FRC purposes, but not because of CIM support. A CIM weighs under three pounds, and it’s only cantilevered out a few inches. If you were worried about it, you could move the CIM back a bit farther with some spacers and put a bearing in the far wall to provide more support than you should ever need for the CIM, unless you were expecting someone to stand on the motor. The drive shaft, on the other hand, can easily support half the weight of the robot as it rocks from the front to rear wheels (or vice versa). For this reason, I endorse keeping the bearing mount low, putting more of the tubing structure above the bearing hole. (Rule 1 that we figured out about drive chassis design is to first consider the stress of keeping the robot off the ground; supporting motors and wheels is minor next to that.)

The lower pocket is clearly larger than the top based on the number of teeth exposed. As others have noted, It must be large enough to allow the gear to be inserted, unless you’re planning to machine the tubing and gear from a solid block of aluminum ;).

So probably using 1/8" stock would be better. If the spacing inside is too little to fit anything in, then perhaps using 2x1.5 (I think this exists on McMaster) stock or something higher would be better. Unfortunately doing this would cause you to most likely lose the ability to integrate the gearbox into a drive rail

Thanks. The bottom pocket is .125" wider than the diameter of the gear, so I should be able to just slide it in.

3D viewer here:https://workbench.grabcad.com/workbench/projects/gc8-CFjo2IDNWgzEcFg8GkyKWCYQy5hLaEExJhO07p7Mt5#/space/gczLctrCPkmGsjfu2z3XF1MF91d6eJfBQ7sPCuomoAPGnm/link/495146
The bolts are a little bit messed up on the online viewer right now for some reason, bear with it.

I love this design for what it’s goals are (low weight, single speed drive with minimal parts).

I don’t think most of the issues raised above are too big of an issue. I’m not sure I buy the thin wall being an issue for mounting the CIMs, and if the wall of the tube really did flex from the weight of the CIMs, some simple part (3D printed would be a good source) that goes between your belly pan and motors would solve that problem easily.

As far as pressing the bearings into the tube for the drive shaft, this is the only thing I could see being an issue. A simple fix would be some small 2D bearing blocks that mount to the tube for the bearings to be pressed into.

I would definitely just incorporate this into the drive tube. I don’t see the benefit of it being a bolt on gearbox; at that point you might as well go with some bolt on COTS gearbox.

Maybe I missed it somewhere in the thread, but what gear is on the drive shaft? I think I saw someone say 64T. I think if you switched that up to 72T, that would probably be a more reasonable ratio for a single speed drive, and it would also easily solve any clearance issues with the CIM mounts and allow you to stick with 2x1 (which I think is necessary for it to be a viable choice to incorporate into the drive tube). A larger drive gear would push the CIMs out, allowing you to rotate the mounting holes a bit, keeping your 1/8" clearance from the gear while gaining some clearance from the top wall of the tube.

Great design, I really love it. I had worked on a similar idea before, but somehow that design ended up migrating away to having the gears on the outside of the tube (in the chassis) with a chain-in-tube setup.

EDIT: You may have had hesitation about a 72T gear on a 4" wheel; that would explain the 64T gear.

EDIT AGAIN: Disregard the part about the larger gear buying you more clearance space, I misunderstood how you had the mounting set up. I sketched everything you have out and see your dilemma.

Love the concept. There are a lot of “you should do X” comments in this thread, but generally speaking this is a great design as-is.

Say your robot comes in at 150 lbs (w/ bumper/battery), and the game is similar to 2012 where typical strategies go about 20 feet in a single sprint. In other words, you’ve determined you need a 2nd stage for that game. 2015, 2014, 2013, and some 2016 bots wouldn’t need a 2nd stage at all, so this is pretty situational.

Since this gearbox is modular, a new gearbox with the same interface should be plausible. How would you re-design the gearbox to do it?

edit: for reference, you’ll want to go way back in JVN’s spreadsheet history:
https://www.chiefdelphi.com/media/papers/2750

Thanks! I was afraid that the gear might get caught in the carpet. A 70t gear might work though. I have no experience with large gears getting caught in the carpet. :confused: How close can the gear get to the ground and still be fine?

I’m expecting a lot of anecdotal answers, since the Vexpro Single Reduction Clamping Gearbox has been out for 2 years now!:stuck_out_tongue: