Awesome robot, 'Ranglers. Can’t wait to see it in person next week!!!
We will be up against a Raptor and a Dragon…and we look like a duck…a mighty duck…
Another in this series, discussing the one of the reasons we chose to build the design we did, and some more detail on the wrist joint:
http://blog.iamjvn.com/2011/02/inside-raptor-philosophy-wrist-joints.html
-John
1 Like
John – thanks for taking the time to do this series of blog posts. I often wonder if the way I conceptualize and design things is “right,” or if I’m off in my own corner of the world somewhere doing weird things that are crazy.
The manner y’all use to extend the upper part of your four-bar is exactly like something I’d conceptualized a few years ago after I was unhappy with the telescoping mechanism on our 2005 robot. It’s awesome to see that someone else had a similar idea and even better to see it implemented so well.
Here is another in this series describing our use of aluminum plate gears this season:
http://blog.iamjvn.com/2011/02/inside-raptor-plate-gears.html
-John
John,
I’m curious how you avoid misalignment of the teeth relative to each plate after assembly. Do you bolt both plates together and then cut the teeth? Do you use pins to align them and then clamp them with bolts? is it just “good enough” that you can bolt them together without any alignment aids?
I’m curious as to why you chose to stack plates instead of just cutting a thicker plate? I can understand if it was really thick for cost purposes, like over a half an inch, but wouldn’t it be easier for a 3/8 piece to just cut one from a thicker piece of metal? I suppose if you’re laser cutting then thicker aluminum isn’t an option on all lasers because of aluminum’s reflective properties.
I’ve also always been curious, are your parts lasered of waterjetted?
It is just “good enough.”
We align the plates (using a mating gear + eyeballs) and then clamp them with c-clamps before riveting.
These are aluminum gears, and we figured they’d wear in happy.
Again – we were surprised how well this worked. We were expecting more hassle… but… it just works.
-John
Our parts are lasered. We use 1/8" plates stacked instead of thicker plates for two main reasons:
-
The laser does have some “blow out” which will affect the tooth profile in thicker materials.
-
Our shop stocks 1/8" 6061 aluminum, so we can get it without a special order. When I’m trying to get 24 hours turns, I try not to make our shop do anything they don’t normally do for production.
-John
Cool stuff, John!
Maybe you guys need to start up a Robowranglers R&D division to test more neat ideas like this… you can attempt all kinds of weird projects that would normally be foolish to try during a build season*. Though it sounds like your offseason work already accomplishes this. 
cough walking robot cough cough
If you dislike chain so much, why did you choose to chain together your omni-wheels and traction wheels? From your models it looks like you have enough CAD experience to space and use timing belts. SolidWorks (at least, maybe Inventor does too) comes with a tool for calculating belt lengths that’s very handy. Belts are quieter and don’t need tensioning.
Maybe next year.
We dislike chains that we need to worry about. This low-load application is pretty happy the way it is.
We’ve used laser cut gears in manipulators before when a small range of motion was desired but Ive never thought about using them in a drive application. Be sure to post up at the end of the season about any wear issues you encountered because otherwise it is a very simple solution to custom gears. I’ll be sure to check them out at alamo.
I wonder if there is any maximum allowance of people/teams per pit. Looks like 148 is going have a crowd. Non-stop. Again.
How do you generate the tooth profiles? Does IFI just have a piece of software that spits it out, do you do it by hand :ahh: , or is there a handy SW feature that I’ve missed that does it for you?
Well 148 would be nagged about the safety issue of too much people crowding your pit.
Actually, it’s always a pleasure to notice how planning committees work to make the pits efficient and, at the same time, a showcase for teams like 148 and 118, allowing space where there is some to be had. It’s also nice when rookies are placed between veteran teams or are placed in an area together where they can be accessed and helped easily by veteran teams and volunteers who will help. Teams like 148 are very aware of their draw and they work hard not to infringe on other teams’ space. I’ve seen it on Championship level and Regional level. It’s also the job of the pit volunteers to help keep things orderly and running smoothly.
There are lots of ways to do this. Some of my favs:
-
Pull out your handy Machinist’s Handbook and sketch the tooth profile by hand.
-
Buy one of those gear profile generation programs and get it to spit it out for you.
-
Download the gear from somewhere online. bostongear.com has all their gears available online.
For this year’s robot we used option 3. I downloaded the gear I wanted from Boston gear, then traced over the tooth in Solidworks (eliminating splines with simple arcs, so it would import cleaner into our laser cutter’s NC program). I grabbed this sketch, dropped it into a new sheetmetal part and away we went…
-John
A few more options:
- It’s surprisingly easy to generate a real involute using equations; this and this explain how to do it in Pro/E (but any CAD software ought to be able to do the same).
- If you can make do with an approximation instead, either follow the second method in the link above, or get GearGen (I think it’s shareware*) from here or here. (It’s a DOS program, so the interface might be a little unfamiliar nowadays…) This makes a sort of segmented polyline when exporting, with arcs that (as near as I can determine) are approximate. But thanks to modern, fast computers, you can create an arbitrarily large gear and scale it down to whatever degree of precision you need—so this isn’t a problem so long as your CAM software won’t choke on the geometry. (Note that all of those facets will make 3-D CAD really slow to regenerate; use the equation in that case, or just approximate it for visual purposes.)
*Does anyone even use that term anymore?
Another from this series which highlights the Drive Module layout.
http://blog.iamjvn.com/2011/02/inside-raptor-drive-module.html
-John
Nice drive module!
for making that strange gear for the arm lifter…you could also get the arm gear from a door window regulator from a car, and trace it onto the aluminum