Rough draft of a drive concept, based on feedback I received on my gearbox idea. The main motivation is that I don’t like the unserviceability of belts in WCD, nor do I like ceding the entire middle of the robot to the motors, so I wanted something that would preserve the simplicity and easy wheel access of a WCD while allowing easy belt access and gearbox removal.
Drive consists of modified WCP 3CIM dog-shifters with CIMs floating over the wheels. 4’’ VexPro traction wheels, power transfer with 9mm VexPro belts. As was suggested, the gearbox shaft couples to the center wheel shaft with a hex sleeve (not visible, enclosed in the 2x1.5 that the gearboxes mount to). Could save a bit of space/weight by changing the 2x1.5 to 2x1, but I’m not sure if the coupler would engage enough of the hex shaft in that small amount of space.
Chassis construction would be with gusset plates (not shown) and 10 bolts or rivets.
In this setup, you don’t need to remove the output shaft of the gearbox at all; the center wheel shaft pulls right out after loosening a single shaft collar. It’s coupled to the output shaft of the gearbox with a hex sleeve.
Another nice feature is that to pull the gearbox, one only needs to slide it about an inch in after removing the bolts, as opposed to the full length of a WCD-style output shaft.
I think this is a great idea, it saves a lot of space for the electronics, but i think you should consider that this makes you CoG higher, increasing the change of your robot tripping. Also the cims are way more exposed out there, and you should consider that. The cables coming out of the cims will also be out there, you should check if they wont leave the frame perimeter or get bent too much. Where they are right now unfortunate accidents could happen, even if the chance of someone messing up your wiring out there is low you shoud be very carfeful with that, some teams tend to have the worst luck during the regionals. At NYC this year our main braker was exposed and a team managed to accidentally hit it in a way that it turned our robot off, so i suggest making a case for the cims and the wires if you keep that design.
If you are really concerned about packaging, I would get rid of the 2x1’s on the outside of wheel well and replace it with .25 in plate, or get rid of it (WCD axle’s are cantilevered; outside support not necessary.)
I don’t know about dead axles but with live axles you can replace the box tubing on the sides with plate tubing. I’ll try to find a photo of it if you wish to see.
In our experience, it works quite well for keeping the COG low and keeping the weight on the outside. The gearbox and CIMs are also very accessible. There’s quite a bit of force on the inner gearbox plates though (the entire gearbox is essentially resting on them), just something to bear in mind.
It is a good start. Here are a few questions though:
Why not get rid of the unnecessary metal on the inner gearbox plates? Leave some meat around the upper bearing and the get rid of the rest.
Why did you place the pulleys on the outside of the wheels? For turning, stability, and other purposes it would be better to get the wheels out the extra ~1" on both sides.
Is your center pulley wide enough to fit 2 belts side-by-side? All the pulleys appear to be the same size.
CG does seem high, but depending on the game this may not matter. If you need essentially no ground clearance (this year), you can have your battery sit about a 1/2" off the ground. [Legally] Max your bumpers out to 20lbs and place them as low as legally allowed.
How are you bolting the chassis and it’s components without collapsing the tube? It is not typically a problem with 1/8" walls, but every now and again you will have someone not paying attention and keep cranking on the bolt because they can still turn it.
I fully intend to do this; this is just a rough draft I banged out in an hour or two. The extent of the gearbox plate design was essentially copy-pasting projected geometry from a WCP 3CIM dog shifter and a WCP 3CIM single-speed into a single sketch and then toying with it until the center-to-center on the gears was correct, and adjusting the outline.
Why did you place the pulleys on the outside of the wheels? For turning, stability, and other purposes it would be better to get the wheels out the extra ~1" on both sides.
Good point, I’ll change that.
Is your center pulley wide enough to fit 2 belts side-by-side? All the pulleys appear to be the same size.
The center pulley can fit two 9mm belts side-by-side. The outer pulleys would be narrower on the actual thing, but to save time while doing this (again, rough draft) I just copy-pasted the center wheel assembly.
How are you bolting the chassis and it’s components without collapsing the tube? It is not typically a problem with 1/8" walls, but every now and again you will have someone not paying attention and keep cranking on the bolt because they can still turn it.
My general solution to this is either a) hover over the students while they assemble it and remind them or b) make wood inserts. I prefer a), since b) adds weight and is a pain. It is 1/8’’ wall + gusset plate thickness, though, and I’m pretty sure I saw someone shear a 1/4-20 bolt head off at 449 putting a similarly-constructed thing together without crunching the assembly.
All good points. I did make sure to leave myself a fair bit of breathing room so that I can keep the wires inside the frame perimeter - I find securing them to the body of the motors with zipties works wonders. I’m not so worried about high CG with this given the low clearance of the chassis above the ground.
While I agree with you in principle, I don’t think it’s a particularly high concern here-- the weight isn’t much higher than it would be with 6" wheels (or for that matter, 8" wheels). There are plenty of gearboxes (including the pre-2014 channel kitbot) which used gearboxes that were on the whole significantly higher up (being above the rail), and I don’t think any teams had issues from that. Overall it’s the other ~70 lbs and batter that you put on it that makes the real difference.
Eli, nice design. I hope you post a more finished version when you’ve cleaned up the CAD and models. Maybe even a pretty render for us?
I hope to see something similar coming out of your team for competition in the future.
We used 3/16" aluminum rivers with steel mandrels to hold our chassis together this year - I don’t think we’re ever going back. They are simple and lightweight.
Another option is Rivet nuts (riv-nuts). They are exactly what they sound like. You can get them from McMaster (94020A335) and other places.
If COG is a concern, you could do an 8 wheel drive and rotate the gearbox 90 degrees provided you move the belts to the inside of the drive rail, and design the mounting plate of the gearbox as a frame element.
I’m really not all that concerned about CG; the highest point of this whole assembly is ~10’’ off the ground.
Re: Pulley sizes, after toying around a bit I’m pretty sure I’m going to keep them all the same size, for ease of replacement. Making the outer pulleys smaller involves either greatly complicating the wheel spacer arrangement (which is very simple right now) or having the wheels out-of-line, and moreover requires that you keep more backups.
