The perfect gearbox
 

If you were designing the perfect (or as close as possible) gearbox (for FRC robot drive system) what features would it have? Here's some basic points but feel free to add more

Selectable ratios Y/N and how many?
What motors input and how many?
What speed (rpm) output?
Size
Weight
Mounting
Output

And just for fun, if you had to buy all the parts and materials to build this perfect gearbox, how much would you spend on it?

Re: The perfect gearbox
 

I wouldn't emphasize on this gearbox, probably just use the one in the KOP and I would worry more about the other functions of the robot.

Re: The perfect gearbox
 

I'm just hypothetically speaking. Say you didn't have to build it. Say it came in the kit. Or say I walked up to your doorstep with a couple in my hands. Regardless of having to actually build it, what features/specifications would it have.

EDIT: Let's please try to make this a useful (non silly/sarcastic) thread. Perhaps we should stick to the main topic points in my first post.

Re: The perfect gearbox
 

Oh well then, I want the whole thing made of GOLD :D, pnuematic 4 gear transmission for no reason. Shifting is not really important, and can get to be pain. Just like driving a manual.. bah. Just make the gear box simple. And if money was no problem then just make it out of gold or silver

Re: The perfect gearbox
 

cheap, light, reliable, shifting is always nice

Re: The perfect gearbox
 

The gearbox can be anything, just as long as it weighs less than 0 lbs, and yes, I'm talking about negative weight here.

Re: The perfect gearbox
 

Light light light... A heavy gearbox is not your friend.. Preferrably 2 speed, anything more than that on an FRC robot right now doesn't seem to have much of a point.. Preferrably having a way making it near impossible to lose a part... That's always a problem... ("Where'd the sprocket go...?") Oh... And looking nice is always a plus. :P

Re: The perfect gearbox
 

Hmm, I've thought about this before and I've come down to a few conclusions. A CVT would probably be awesome beyond belief, however its quite heavy, not to mention if it really was a CVT it wouldn't be a true CVT.

So, that being said I do have another idea thats a bit more realistic:
3 speed, 1-Mini Bike, 1-Cim in combination. The smaller cim would be geared down. Next I would make it something like a 3-speed plantetary with a servo powered, ball bearing shifter. To add some kicks i'd have encoders on it so it could be programmed to drive like an automatic.

In all, the gearbox without the motors would weigh less than 5 pounds (probably 3ish), and I'd have two total. I wouldn't try anything too special like a holonomic drive system or mechanum wheels, because i watch teams use them and the driver never seems like they have full control of the robot. I want something thats easy to control and with good handling. The most unique I'd get with this design is that because its a planetary and servo shifting, it'd be small enough to use with a swerve drive, yes imagine a multi-speed swerve drive.

Hmmm, I got to get to work on this.

Re: The perfect gearbox
 

I would like a 2 speed shifter, which could be directly driven to a 4" wheel.

I'm not sure on the gearing, but I would like the actual output to be 12 ft/s high and 6 ft/s low.

There should only be one reduction done after the initial motor combiner, and all shafts should be riding on ball bearings. I've definitely seen the adverse effects of loss in efficiency. This year's transmission with 2 less reduction stages (last year 4 speed, this year 2 speed) has done wonders, even at the same output spec.

But, if I had to pick something that is easily accessible off of these boards, I would 100%, without a doubt go for the team 33 style blind shifter. These things are absolutely beautiful, and NEVER have to be maintenanced.

Re: The perfect gearbox
 

I like my drive train this year. 7 fps low, 14 fps high. Torque and speed, and a nice 6 lbs flat with the shift piston and motors included. The CIM and FP give a good bit of power. Construction out of 1/4" aluminum plate with pressed plain bearings, 1/2" OD threaded aluminum spacers, hardened allen head cap screws, and a very narrow, large OD dog shifter. Makes for a very small (1.75" wide), very light transmission.

Personally I would always stick to 2 speed dog shifter made of 1/4" plate. It is a very solid base to work off, and my teams have had lots of experience with this kind of transmission. Of course we always work for new advancements every year, but this is our sure proof drive train.

NOTE: New features of this year's transmission are: quick 4 bolt face mounting, pneumatics mounted directly on the transmission (as opposed to chassis mounted), larger diameter dog (1.375"), narrower transmission (by 1/4"), more efficient (can anyone say 'valve grinding compound'?). Not a whole lot, but sure makes maintaining it a lot better.

Re: The perfect gearbox
 

may i sugest using a gear drive train. I think it might have a chance of working well for you

shaun

Re: The perfect gearbox
 

I always liked the ball shift gear box that team 222 uses. Never liked the idea of smashing gear teeth together.

Re: The perfect gearbox
 

CVT YES i was gonna put it but you beat me to it that would be the icing on the cake.

Re: The perfect gearbox
 

I have heard wonderful things about the dewalt gearboxes. Three speed, servo shifted, light, small, virtually indestructable, cheap. If FIRST wanted to include a powerhouse gearbox in the kit (Which I doubt they would, unfortunately), I would be very happy to get a couple of prepared dewalts. Wait, no, ecstatic.

Re: The perfect gearbox
 

So basically it would be a 2 speed shift on the fly transmission. The motors powering it would of course be the 2 Chips (and possibly the minibike). The arrangment would be that of a triangle, with the 2 chips at the bottom and the minibike at the top.

The gearbox would be geared for 11ft/sec (633.75 rpms using 4 inch wheels) high, and 5 ft/sec (297.99 using 4 inch wheels) low.

I would give around 8 pounds for each of the transmissions, but then again it is like the most important part of the robot, so if it needs to be a little heavier she can be.

Assuming that we are given the same chassis as we have been given for the past 2 years it would sit down within the chasis, with 2 output sprockets running to each end of the drive train.

Re: The perfect gearbox
 

While I like the 2:1 or 2.5:1 ratios that most gearboxes provide, I REALLY like the 3-speed ratios that the Dewalt's provide. Using the CIM, it provides 3:1, 4:1 and 12:1 ratios, and only requires the use of one motor. The "stump puller" ratio, as I like to call it, gives plenty of torque with a one motor setup. (Granted, at the sacrifice of a bit of speed.) While the other two ratios provide plenty of speed to get across a wide open field, a la this 2006 game...

Overall, the planetary setups provide the smallest, lightest packages possible. Let's face it, 3/8" face, 50 tooth gears are never going to be the lightest for a FIRST robot. Look at the success of the Banebot's gearboxes this year. Cheap, light, small, and a lot of people are using them, even though their quality may be a bit questionable... Make a similar gearbox for a drivetrain (with a bit more robustness) and I think you've got a winner.

My only gripe about most gearboxes is that they take up a lot of volume. Specifically, they tend to be quite wide. Folks using an off-the-shelf gearbox for their onmi-drive setups, or even just a standard 4wd find that the overall length of the motor + gearbox setup becomes very long. Andy Brockway and the 229 crew have solved this problem nicely with their gearboxes. Maybe something like that can be integrated into a planetary setup.

Also, if you look at Wildstang's, Simbots, or even Beatty's (and I'm sure others') custom drive gearboxes on all their swerve setups, you'll find a beautiful, lightweight package. Let's see the whitepapers on those bad-boys!

I believe the kit gearbox is brilliant, and why some under funded, under manned teams shy away from it is beyond my grasp. It's a tough package to beat for reliability and ease of use. You need to match it's performance and ease of use if you want to sell into this group.

Though, overall, I think a lot of work can be done to help teams integrate their motors into a nice, off-the-shelf gearbox. The motor rpm's this year are either really fast, or really slow, and teams are finding that it's tough to get something running in the 500 to 1000rpm range with a small, light package. Adapter plates, mounting brackets, etc... take the pain away from teams with just a drill press and a hack saw.

I know this got a little off-topic, but I think making things easier for the smaller teams will greatly increase the competitiveness of all the robots. If you can make it easier, AND add a performance increase (through a shifting transmission or whatnot) you'll find yourself with a winning package.

BEN

Re: The perfect gearbox
 

Thanks, we built another one this year and its better than ever!

We have not released our 2006 transmissions yet, so here is a sneak peek, but I think we came close to what this thread is asking!

They are smaller than the KOP transmissions, have two speeds, easily changeable gear ratios, can use one or two CIMS per transmission, and they weigh less than the KOP transmissions.

See them at VCU!

Re: The perfect gearbox
 

As some have requested, here are some pics of the 2006 HRT Team 1072 drivetrain I described earlier.

Re: The perfect gearbox
 

I have a few dream 'boxes for different purposes.

General use: A two speed dog (Woburn style) with aluminum gears (more on these later) and a polycarb/aluminum shell. Now to explain my odd materials choices, the polycarb would be ultra thin, whereas the aluminum would be beefy. Now the aluminum would make a crisscross structure supporting only the high load points (eg the bearing mounts) and forces (eg the mounting holes) on the gearbox. The polycarb would mostly to keep garbage out, it could even be saran-wrap. The aluminum gears wouldn't be strong enough especially considering they'd be milled for lightness, except for some tricky chemistry. Drop them in a vacuum chamber with H2 (hydrogen) and CH4 (methane) at 50 Torr, add heat and voila! diamond coating. Or you could just pay a company to do it, whatever.

Insane use: A Thunderchicken CCT with the aforementioned polycarb/aluminum structure and lightened diamond plated aluminum gears. It would be driven by the bike chip, and two small chips with the FP providing the ratio control function.

Insane use 2: This box also uses polycarb/aluminum/diamond coating construction. It uses a ball bearing shifter with three planetary gear selections. It is powered by the bike chip, two small chips, and the FP (maybe a Globe and window motor too!) Bring on the twelve motor drive!

Absolutely Insane use: Okay, time for my nuttiest idea yet. There is a central gearbox that matches ratios between all the motors in the kit (except that crummy Mabuchi motor) to, say 500 rpm. There are then two separate sets of CVT v-belt assemblies. These allow for the speed on one side to be reduced to, say 50 rpm, with the other side doing 300 rpm allowing for skid steer turning. This could also have four CVT assemblies for holonomic. Yay! as many motor as you have drive!

It's times like this when I wish we had more money and machining capability.

Re: The perfect gearbox
 

hey
are the dewalt gearboxes compatible w/ the CIMs?

Re: The perfect gearbox
 

Yes, with some modifications. See the Nothing But Dewalts whitepaper for details.