So after reviewing a few different drive-trains that were posted on here I was curious about packaging a gearbox into frame. The attached was the result, and it has been dubbed the ‘Frame Drive’.
Some advantages:
-Low center of gravity
-The ability to machine the components on a hand mill and lathe.
-Gearbox and frame integrated
-Single speed (no shifting complexity)
-Low maintenance (mainly greasing gears)
Questions and comments are welcome!
Hopefully this can provide some inspiration as other postings have inspired as well.
I’ve been daydreaming about this concept also. My stumbling block has been the insanely high speed – maybe 20+ fps assuming 1x3 frame channel and 4" wheels.
Last year my team’s drivetrain was geared 7:1 with 4" wheels and 3 CIMs per side. I would not want to drive anything geared faster, so maybe it is a good thing I am too old to be a driver.
The tubes are 3x1 with 4 inch wheels, it is geared 6.94:1 at the moment if i did the math right. 12 tooth to 50 tooth to 30 tooth gears with 1:1 belt drive on the outside wheels.
I think the main issue would be ground clearance with the 3" tubes and 4" wheels.
Unless I’m looking at the picture wrong it looks like all of your gears on the same line, that won’t yield you two reductions like you think it will. It’s actually just 30:12. You have a 12t gear spinning the 50t and then the 50t gear is directly spinning the 30t gear.
As Allen pointed out, inline gears, like you have, don’t stack up ratios. The gear in the middle doesn’t do anything when calculating the speed, so you really have a 30:12. If you want to stack up gear ratios, you have to do something like:
Also, if you are concerned about ground clearance, the wheels don’t have to be exactly in the center of the tube. Move them down if you need to.
And remember that you are going to need bumper mounts, but that should be too difficult to add in.
Nonetheless, this is a pretty cool concept, and it leaves a lot of space for electronics. Add in bumper mounts and possibly a bellypan, and fix the gear ratios, and you could have a pretty cool chassis.
The shafts were placed in the center of the tubes for gear clearance, but with the incorrect ratios used from the above posts, things could be shifted. The first thing that comes to mind are a few gears external to the frame to give the correct ratio.
I’ve been playing around with similar concepts, and yours is definitely a cool way to tackle the problem. I’m not sure what your other design constraints are, but if I were you, I’d look into moving the final stage of gearing (I’m assuming you’ll stick with the two stages) to the outside of the tube (between the two center wheels), an adding a gear onto both center wheels on either side of the drive. This should free up some of your gearing options on the second stage, and I’d imagine you could bump the final gear all the way up to a 60(or 64/66 depending on how much clearance you can live with) to get a bit more reduction going. There’s also the added bonus of not necessarily needing live axles on the center (4) wheels - or any wheels for that matter - meaning that you might end up netting a bit of weight savings while you’re at it.
It’s also worth nothing that VP will be selling an 11 Tooth Pinion (and 13T) during the 2014 Season, and they’re kind of Magic when it comes to squeezing a little extra reduction out of a tight box.
If I was putting gears in frame rails, I’d copy some elements of AndyMark’s assorted Nano Tube products. They use 1.5"x3" tube to fit two gear stages inside the tube. If I was doing that, I’d probably either run chains inside the tubes (easier with the bigger tube) or do a full gear drive. Then I don’t give away the space taken up by the pulleys between the wheels and the rails.
I have seen the 11t, 13t and 14t cim gears, and it does seem like a nice way to get a quick reduction. I believe the 13t and 14t have a different c-c distance than the 11t or 12t. The 30t gear can also be reduced for slower speeds, the 50t gear is pretty much set for a 3 inch tall tube and the cim motors. Any larger (unless it is external) it will rub the walls, any smaller than 50t, there isnt access to the backside of the shaft between the 2.5" cim motors. As Pault suggested, the tubes could be cut for more clearance.
I like Nemo’s suggestion of larger tubes, versus an external gear. Protection the gears from potential damage during a season seems valuable. It appears 3x2" tubing is available in 6061, whereas 3x1.5" is only available in 6063. Staying away from 6063 would be nice, and the 3x2 gives more room for bearings and a larger mounting surface for mechanisms.
I agree Nemo that chain will package nicer, but all chain I have dealt with requires tensioner’s because of wear. Although belts are wider and don’t package as nice, they should require much less maintenance.
Ha I guess the grammar is a bit off. You can call it that if you want. The thinking was along the lines of tank drive, butterfly drive, west coast drive, mecanum drive, etc. So frame would be the adjective and drive the noun.
Are you planning on riveting it together? If so, do you have enough clearance for the river gun on the inside corners of the box tube (the 90-degree bracket with 8 holes)?
Did you do any calculations to determine the wheel layout (VersaWheel vs. VersaWheel DT)?
We do plan on riveting it together, and I agree that the corner brackets are very tight. We may be able to get at them with the hand rivet tool, definitely not with the air riveter.
There wasn’t really any calculations done for wheel layout as far as versa W and DT tread go, more of a thought process. Since the middle 4 wheels are doing a majority of the work, the DT tread was designed by Vexpro to have better wear characteristics than the W tread. People had wear issues with the W tread last year. With very little work being done with the outside wheels, they should last longer and give better traction during a sticky situation. Now this is not geared to push, it should be able to cook the breakers for at least 4 seconds to get out of situations. When pushing into another robot the base should rock back or forward to get the better coefficient of friction of the W treads (1.2 versus 1.0 according to Vexpro) into action. Otherwise the outside wheels are useless. So maintenance and pushing were the considerations for the wheels. A 6wd may come to different conclusions since the middle wheel is in action 100% of the time.
I was able to find 6061 T6 aluminum in the 3x1.5" variety, and the gears packaged quite nicely with the 1.5" wide tube. A few delrin spacers should finish the gearbox off.
I really like gears in the tube, but it adds two more shafts.
Prior to the vex 2014 product unveil I was playing around to see how compact I could make a 8wd. I came up with something very similar. It would be super light without the colson wheels, but I have a hard time giving them up. I like to think of this as colsons with cleats.
One of the design constraints was cost, I wanted to avoid using hex bearings as much as possible, and actually used halved 42 tooth AM kit bot pulleys for uber cheapness.
Half live axle, half dead axle, It uses a single width of 9mm belt to power the outer wheels on the outside of the frame, while running the inside wheels on two(yes, two!) direct driven shafts. With the single 12:60 reduction it is geared for 15fps, which even on 6 cims is too fast. a 11 tooth pinion would make it a slightly more manageable 13.75 fps.
Fantastic! I like how two different approaches end up in similar designs.
A comment about the hex bearings that is my opinion and I am sure will differ from person to person. Even though they cost a little more, they simplify assembly by removing the need for precision machining of shafts to get a good fit into the bearing. Round shafts would need a light press fit, or loctite, or maybe some sort of knurl to fit tight into the bearing. So the hex removes the need for any of those three things while providing slip fit and proper bearing usage. It’s the never ending battle of cost versus benefit.
Last year my team integrated our gearboxes into our frame to save space in the middle of our robot for disc movement. We had 1.5"x3"x.125" aluminum tubing, and our drivetrain was west coast style. We took the gears and hardware from a Toughbox and built them into the tube, mounting the CIMs and some bearings to a single plate that was bolted onto the inside of the frame. The design worked and saved space, but we had some efficiency loss because the sides of the frame rail were not perfectly square, and it affected our gear spacing slightly. Also, maintenence was difficult on this design because there was no room to pull the CIMs off of the gearbox- however, this was due to other packaging in the robot, not the gearbox design. IMO if you can package this gearbox with the rest of your robot and you account for tolerance in the walls of the aluminum frame, this could work great.
