Hi everyone I was assigned to cad bumpers for the KOP chassis and I’m still pretty new to CAD we are using fusion 360 and I was wondering if anyone had pointers?
I’m not an expert with Fusion 360 but when we do our bumpers:
- We do a top down sketch for the 3/4" plywood
- Extrude the pieces to 5 inches tall
- Draw a sketch on the end of one of the pieces of plywood with the 2.5" circles for the pool noodles with the outer profile being in the shape of a D.
- Use a sweep command to wrap the bumper profile around the plywood frame.
- Add a fillet at the corners to clean up the look
Here is a link to a quick sample Onshape document: Onshape
Make sure to provide enough clearance between the chassis and bumpers. This is the biggest mistake our students have made in the past. You definitely need some space for the thickness of the fabric, but also a little more to make changes quick and easy.
We have been struggling with this the past few seasons, making bumpers way too tight, so I’m actually curious how much clearance the veteran teams have settled on.
100 experimented for a while and found 1/16" on each side (so 1/8" total) to be good for a tight fit, especially for bumpers that slid in from the side. 1/8" on each side was also acceptable, and was usually better for bumpers that dropped in from the top.
A couple of points to keep in mind:
Allowing 1/8 - 3/16" clearance between robot frame and bumper plywood allows sufficient room to overlap fabric for a clean appearance.
Place a premium on robust bumper mounting system. Angle aluminum mounts securely mounted to the bumpers can be easily un/bolted to robot frame elements. I am not a fan of quick release pins or similar attachment methods. If designing a bumper with openings, the importance of secure bumper mounts escalates.
If a student can yank on the edge of an open bumper and it starts falling apart, it’s not going to last a regional. The number of time I’ve wanted to test that while pit scouting…
Assuming you’re getting started now, and you want to be a team with really good (and not just sufficient) bumpers, there are a few things I would suggest. This might be overkill, or it might be a good way to save your team a lot of frustration at the competitions, when your bumpers just work. These are mainly ideas for implementing bumpers as a system (and not the CAD specifically), but by being able to depict these things in your CAD before the season, you will increase the chances that they will make it on the robot for real.
Check last year’s rules to be sure of the critical measurements for the robot frame perimeter, the bumpers and the bumper zone. Be ready for these to change in 2022 (so compare the new rules when they come out), but historically FIRST has retained most specifications from year to year. Understand what every part of section 9.5 means.
Take note of facts like not all plywood that is sold as 3/4 in is actually exactly 0.750 in thick, not all wood is fantastically flat, and not all pool noodles are Ø2.5 in in diameter. This should prompt a question about whether your team will be able to guarantee exact sizes by buying specific materials, or if you might need to plan your tolerances to work around this uncertainty. Think about your team’s manufacturing capabilities when you decide. If you are able to make choices now, then enter the dimensions of the chosen materials into your CAD so that you reduce the uncertainty.
Think about how precisely you will be able to line up your mounting features, how strong these mounting features will be, and how fast they will be to install and remove. My guideline for bumpers is that each segment should be removable by 1 person in 10 s, and that the minimum set of tools needed (if any) should be easily carried in a pocket (hopefully the pocket of a person at the field with the robot). The fasteners you take on and off should be reusable (don’t use wood screws for this), and it is even better if they are captive (so that they can’t fall in the robot and get lost).
You might want to use the metal clamping angle allowed by the rules (R24-G), because this also provides a good place (on the vertical back surface) to attach very strong, precise mounting brackets. You can build the bumper cushions and backing, and then mark locations to permanently mount the clamping angles with many short, strong countersunk wood screws. This way, the bumper is mostly complete by very early in the build period, and you can work with the metal angle up against your frame to mark out the exact mounting location without having to dig into the cushions to reach the other sides of the fasteners. The variety of materials available as metal angle is large, and many will fit within (or be machinable to fit within) the 1 in hard-part zone (R25) of the bumpers.
Think about whether you want strong bumpers (Baltic birch plywood + 7075 aluminum clamping angle would be an expensive way to do it) or light bumpers (pine boards, and no clamping angle), and to what extent you want the bumpers and frame to do double duty supporting the other.
Remember that the consequence of the bumpers falling off or breaking catastrophically is often that your robot will be disabled, and you may lose the match as a result, so there’s a lot riding on the design being well executed—but by thinking about it now, you’re increasing your chances of success!
I would strongly advise against using boards for your bumpers. One year, the team my younger son was on used oak boards for their bumpers. After a few impacts, they split along the grain. The splits started at the screws used to hold the corner brackets on. I am pretty sure they pre-drilled the screw holes.
When purchasing plywood, one will find different grades. The stronger ones generally have more layers or plys.
Baltic birch plywood is the only plywood I recommend. No boards, no osb, no nonsense.
In contrast there’s no reason to splurge on 7075 for angle brackets. Any piece of 1/8" aluminum will do fine.
We run a tight fit continuous 1-piece baltic birch plywood frame (3/32" per side clearance to allow for fabric*, notched (interlocked) with 4 pre-drilled screws per corner), and get away with only having 4 brackets total.
If you have a bumper opening or especially multi-part bumpers, you’ll need more (and maybe stronger) brackets to deal with prying forces.
*(Don’t go over 1/8" per side. If you miss wide and bolt one side flush, that gives the other end a full 1/4" along the whole frame, and inspectors will be able to make you miss practice matches to add metal until it’s fully supported.)
Here’s an inexpensive COTS solution. They are wider than the aluminum angle that most people use so half of the screws end up further from the end of the pieces of plywood. We have used these for years and they hold up well, even in the years with the most abusive on-field action.
Seconding this recomendation for COTS angle for your bumper assembly. We went this route in 2020 after being inspired by 118’s 2019 Everybot guide (specifically the section on bumpers).
While our bumper mounting system SUCKED that year, the bumpers themselves held up great and these brackets from homedepot worked wonders.
As a clarification, the Simpson Strong-Tie brackets in that size are good for the bumper mounting (hanging over the frame), but the flange length on one side would need to be cut down and deburred if used for the fabric clamping (when hanging over the bumper wood, the 1 in hard parts rule applies).
For mounting the bumpers, Simpson Strong-Tie brackets are good, cheap and widely available—but surprisingly, they aren’t as strong as might be expected. The A23 bracket is 18 gauge carbon steel, and so that means Simpson will only specify it at 33 000 lb/in2 yield strength (thickness 0.049 in). This is in the ballpark of regular 6061 aluminum angle.
Lots of good advice here, from people who’ve learned, often the hard way, what doesn’t work well — and developed solutions that do.
Interesting how fast this topic evolved from the OP’s question about how to CAD bumpers (addressed very well in the first reply) into a discussion on how to MAKE bumpers.
BTW, I agree with those who emphasized the need for fabric clearance between plywood and frame metal. Target 1/8 inch per side. Tolerance on that should be 1/32 inch; i.e., anything between 3/32” and 5/32” will work. Measure carefully and be sure to allow for the actual thickness of your plywood when calculating cut lengths.
We used them in the corners between bumper sections to hold adjacent bumper sections together. We have also installed them on the top of the frame and attached bumpers using bolts and t-nuts. They have been plenty strong for all those applications.
If you CAD good-looking bumpers, but your real bumpers are really badly made and attached, how much good did the CAD do you? We want the good CAD to translate into a good IRL bumper.
Actually, one of my favorite robots this last season would occasionally drop a bumper section onto the field. The catch was that they still had a fully legal and fully compliant bumper otherwise–the section was a middle section in between two corners. Without it they just couldn’t intake very well.
Wouldn’t that still run afoul of G20, and possibly G19?
Definitely not G19–it was almost always a direct contact knocking it off. If we call that intentional then we’ll have to call everybody breaking themselves intentional.
G20 is a MUCH tougher call. You want it called by the letter of the rule, it’s a problem (a full segment was completely detached). But, if you want it called by the letter of the rule, let’s just say that the letter of the rule can get a lot of people in trouble*. The SPIRIT of the rule, as I read it, is essentially “Don’t be playing without bumpers on the robot that correctly identify you”. Because the team still had a full set of legal bumpers, and (more importantly) this was mainly at offseason events, let 'em play. Next year’s regionals, that’s a bit more dicey. Might be worth a Q&A for 2022 if it isn’t addressed in the rules.
*Example: Find me a rule where it specifically says you can use the wrong color of bumper during practice matches. Now find me a team that has been disabled for that offense.