I’m happy to finally be sharing my first real shifting gearbox. In case your summer CD constant refresh has been as boring as mine, feel free to give criticism, potential improvements, or any other feedback here. I’m still pretty new to this gearbox stuff, so thanks in advance for the advice of the people on this “wonderful” site who are much much smarter than I.
Basically, I set out to create a gearbox similar to the WCP Flipped Cim with a couple bells and whistles. For the most part the size is the same, a little bigger on one end and smaller on the other, but I’m most proud of the dog running through another plate (mid-cleaning a 3d print at work epiphany) and the variability within the same gearbox to use anywhere from 11-14t pinions for some tweaking in speed as needed without the use of snowman holes.
Here is the Onshape link: (steer clear of my grossly not organized folder where I threw 3 revisions of junk to keep the main page clear)
And a google drive folder with STEPs
There are 2 versions, one says CIT (Chain in tube) and the other says CIG (Chain in gearbox). Obviously the chain in gearbox one is a bit bigger, but I have some weird ideas for how to get that size down in a coming iteration I may forget to do in a few days.
The gearbox hasn’t been lightened yet because I’m sure someone here will realize a stupid design mistake or two of mine, but with handy dandy featurescript lightening that will take .1836 seconds once the design is locked down.
As always, your questions, comments, and other feedback is greatly appreciated!
I’m at my old computer right now so I can’t open the CAD, so sorry if the answer is obvious in the CAD.
What in your mind are the benefits to being able to switch from 11-14t pinions on a custom gearbox? Since you would have to take apart the gearbox to switch anyway, why not just make 2 different plates (one for 11-12 and one for 13-14) and switch the plates? That would save you a good chunk of weight.
What are the final gear ratios/speeds?
Is there any way to get rid of the idlers using bigger gears and bringing everything closer together? Personally, I’m anti-idler if possible because they add inefficiency, weight, and backlash. Obviously if it is possible to get rid of them you would have to make some compromises, so it depends where your priorities are.
I would be thinking about your frame perimeter and where your bumpers mount. Right now, it looks like the CIM overhangs the tube by > 3 inches, which you may find to be rather annoying.
The cim hangs over 2.554" in the main design, and substantially less in the chain in gearbox revision. I thought that was nice for 3" wheel wells on 2" wide wheels (1.5 pictured cause lazy) which is pretty common with WCD’s, especially on my team. With solid bumpers all the way around, I don’t expect any problems here.
The tradeoff for the idler gear moving around was essentially do I lighten in that spot, or put a hole which is ligntening in some sense anyways. Deep in the rev 1/2 folders there are assemblies without that, I just thought it was a cool concept from the wcp flipped cim that might make things more convenient in the future. One more hole in each part (sacraficing a bit of weight) vs 4 new plates +spares of each? I’ll just add the hole.
I could have gotten rid of the idler by bringing the cims closer together. However then I would not have had room for the actuator in between cims. I could have used belts maybe, but the idler just seemed to be a simple solution so I could still easily and robustly achieve a good reduction
Stupid me forgot to post the output speeds. When i get on my laptop tonight I’ll recalculate and post.
That’s a lot of wasted room, bud. 1" on either side is a pretty substantial reduction of your frame perimeters. Usually I try and make my wheel assemblies less than 1/2" from the frame perimeter at most, and even then I try and get 1/8" or 1/4". My last one had 1.5" wheel wells for wheel assemblies ~1.3" wide. The WCP DS Flipped is designed to fit any size wheel, which require the motors and the shifting cylinder to remain within the tube limits. If you’re designing for a specific wheel in mind you can move everything out and contour the motors to the wheel shape (which in a way you started to do). Try moving everything closer to the wheels since it looks like you’re pushing your motors up higher and out further than they need to be, and eliminate excess overhang of the motors and you should be gucci.
I’m digging the arrangement of the shifting mechanism - keep it up and make sure to share futures iterations!
I guess in that configuration 1" could seem like a lot, but with bearing blocks and all that stuff adding up it’s not too bad. I have also seen teams run one or two chains on the outside of their drive rails, so maybe the space can be used there.
Do keep in mind though that I posted 2 gearboxes here, and the other one has the Cim sticking a little more than .75 out from the tube. That one has the chain running outside of the tube, so it is a little bigger inside the tube but it takes up less space outside. I guess the tradeoff here is bulk for the width of the gearbox, or bulk for the entire robot length, so bulk for the width of the gearbox wins all things equal.
Definitely going to experiment with pushing motors closer to the wheels (maybe I can get rid of the idler there), or even push to 3.25" wheels which I know some teams have had some success with in the past.
Of course I’m sharing future iterations! Can’t rely on asking Adam for all of my design reviews. Don’t suppose you want into the 1836 slack too… It’s a party with all the other teams mentors we have in there. There’s also that fun rule that says you have to share stuff online to use it during season, so rather than sneaking it into some random page deep in milkenknights.com with some of our old stuff posting it here helps get community feedback and entertain those who love the boredom of summer CD just as much as I do.
Thanks for your feedback (and lolz cause bud’s bodyboards)
Yeah you can, would be a helluva lot more valuable than 90% of the advice you find here. It’s why I stopped asking for design advice on chief - you’re better off just privately messaging the powerhouse designers/mentors you have contact with and getting their feedback. I’d try and get a few different points of view for everything, but CD just doesn’t have the quality assurance in terms of technical advice that those established mentors have. You may find a right answer but it’ll be hidden under a stack of responses from bozos like me who think they know how to build a good robot but haven’t actually done so themselves. :rolleyes:
That doesn’t mean don’t post your stuff up here, just be selective about who you take advice frp, and use CD as a platform of showing off your projects instead of seeking help on them.
