Good bumpers are hard. How do we make them easier?

Title says it all. How do we enable building good bumpers and make it easier on teams to do it?

25 Likes

I think FRC is too easy. We need to add hard components, not make them easier!

27 Likes

I’ve been saying it for years, whoever starts selling COTS bumpers for $400 a piece will make millions!!

20 Likes

Provide every team a template for the corners of the fabric.

16 Likes

This is a great question and one I, Al, and every LRI and RI I know would love to have the answer to. But the bigger question I have is WHY are bumpers so hard? If we could firmly understand why teams struggle with them as much as they clearly do then maybe we can begin to find solutions to the issues. There are detailed diagrams in the Section 9 Robot Construction rules. What exactly about the bumper rules and diagrams do teams find difficult? We have discussed all sorts of ways to hep teams better understand them including separate training documents, how-to construction videos, etc. but before any of those efforts can begin to be fruitful a firm grasp of the problems is necessary. As RIs, we see the seemingly never ending issues but exactly what the root causes of those are seems to be a mystery. Where is additional clarity needed in the bumper rules and diagrams? Some have suggested replacing the diagrams that have been more or less the same for a decade or more with example photos. I’m not convinced that photos would cover all the bases and issues and provide any more clarity than the diagrams. Help me understand WHY teams seem to continue to struggle with bumpers so I can endeavor to add clarity to the rules set with the aim of providing useful guidance in some useful way. The last thing anyone wants is MORE information. More documents to read. More how-to videos. But I do wonder if bumpers are something that might warrant more clarity or guidance to make a better inspection experience for teams.

One of the main observations I have had over the years is that a lot of teams sort of leave bumpers as an after thought. Something to do “later” rather than include in their design from the beginning. How do they attach? How do the bumpers alter the interaction with game pieces or the field? Many teams simply underestimate the importance of bumpers and, to be honest, a lot of the issues we see and Robot Inspectors stem from teams that think they can wait until they get to their first event to build their bumpers there. That is typically a underestimation of what is needed to build good, rules compliant bumpers and at a minimum it takes away resources that are probably better spent doing something else more productive, like playing on the practice field, etc. Why do teams underestimate what it takes to make good bumpers? Are they really that hard or are a lot of the team’s issues not so much related to the difficulty of making “good bumpers” or are a lot of the issues underestimation of the importance of including bumpers in the robot design and fabrication early in the build season?

17 Likes

$400/set x 2 sets x 4000 teams = $3,200,000

checks out

And the leading causes of bumpers being difficult can likely be summed up as “the exact opposite of building the rest of the robot”

Fabric, pool noodles, and wood to an extent are not dimensionally stable, fabric is very rare on FRC robots, pretty much all the tools you need are different, getting numbers onto fabric has a number of traps (just stencil and paint them), people introduce complications like reversible bumpers to the conversation, etc etc.
They’re weird and nobody focuses much effort on them because they’re only a small part of the robot. I cannot count the number of offseason robots where bumpers are intentionally handwaved and not taught about.

24 Likes

To add some concrete examples to Chuck’s question…

One year, I had a team build their bumpers using lumber that measured 1"x4".

One year, I had a team iron their bumper numbers on after assembling the bumpers, melting halfway through the pool noodles.

One year, I had a team take incredibly large pool noodles and cut two sides off them to make ones that were approximately the right volume when it was all assembled, as they couldn’t find the right size locally.

One year, I had a team show up without any bumpers.

And that ignores the more common “not mounted securely” or “wood wasn’t cut to the right length” issues that we see at every event. I would love to hear from teams that have these bumper issues - why was it so hard?

8 Likes

Our team is working on building an offseason robot, and I already know getting the bumpers made for it will likely be the most annoying part of the entire process mechanically.

Troy pretty much summed it up; not dimensionally stable, fabrics is annoying, most every robot frame is different so mounting is not consistent, etc etc etc

7 Likes

But if bumpers were easy, how would we know the good teams to pick? /s

27 Likes

$5 says that teams don’t treat them as a priority, and put them off to the last minute. As for why, they’re like the IT department at a company: You don’t need them until you NEED THEM NOW, so of course you can cut a few corners, right?

24 Likes

You could always end up like us last year. We always have great bumpers since we have a professional seamstress as one of the mentors but back to last season. We finished them up pretty early then the practice before our fist competition we noticed the some of the plywood warped really bad and was over an inch away from part of the frame so we had to last minute run to home depot by a new sheet of plywood and remake our bumpers in a couple hours. We got them done but it was a rush we weren’t used to.

1 Like

On the products angle piece: what if there was a dimensionally and functionally equivalent COTS product to the bumper cross section (e.g. a hard backing with foam cushion, and protective outer wrap surface), so all teams had to do was cut it to length with a hacksaw and stick on numbers? It might not be made out of wood, pool noodles, and fabric (for ease of mass production) so rule changes would likely be required, but it would certainly ease the production side. I’m imagining something like a thin hard plastic shell with foam infill, but other ideas of course would work too. Figuring out easy, flexible mounting (e.g. of window latches for quick change) with such a design is the other challenge, but we have a lot of smart people here…

9 Likes

I love making bumpers- but even for myself it’s the last thing to get done on the robot (usually the day before our first event).

My biggest gripe is that our bumpers can pass inspection at one event, and not at another. Many times I’ve had to show the rule that says “5 in. ± ½ in. tall
plywood” when ours are 4.5-4.75". I’ll never forget an inspector having students hot glue 1/8" spacers to our robot frame to take up the small gap where the plywood had warped between events.

In my experience the bumper rules can be difficult to follow for inexperienced teams. There’s 6.5 pages in the 2022 manual to explain plywood and pool noodles. I think a good start is clarifying the rules with actual photos, or a how-to basic document on bumper fabrication. Is there a video similar to the field walk-around videos that shows examples of what does or doesn’t pass inspection for bumpers?

10 Likes

But the thing is they really aren’t. Next to the actual drive train that they attach to that is more or less a requirement, good rules compliant bumpers are probably the next most important part of the robot. You can’t get on the field without them. With a kitbot and bumpers you can at least get on the field and drive around and do SOMETHING. So it seems to me this logic that bumpers are a small “unimportant” part of the robot that can be left to the last minute is one of the main root causes of why teams struggle with bumpers. So, to me, the real question we need to be asking isn’t why are bumpers so hard, because they really aren’t folks. If we can all build very complex robots with all manor of mechanical systems, wire them up, program them, and make them do crazy cool stuff like pickup balls and shoot them through holes in the wall or climb monkey bars I have a hard time believing that cutting some wood strips 5" wide and wrapping them with some 2.5" pool noodles and durable cloth and slapping some legible numbers on them is out of the realm of capability for even the most resource limited teams. The real question we need to be asking is not why they are “so hard”, but why do teams think they are “weird” and not focus much effort on them because they don’t think they are important?

9 Likes

I think one of the reasons why bumpers end up being hard is that there are a ton of very specific rules about their construction and mounting that don’t necessarily align with the intent of requiring bumpers.

Bumpers exist to soften collisions between robots and the field. They need to stay on for the whole match and continue to do their job between matches. They also conveniently double as visual representation of both alliance color and team number.

I think if we took a closer look at the 6.5 pages of the game manual dedicated to bumper construction, we could both reduce the ruleset and improve the messaging on bumper intent.

23 Likes

This is the kind of feedback I am in need of. Please help me by taking a stab at those 6.5 pages of bumper rules and let me know which are superlative, which are unclear, and how you might reword them for more clarity. This is my open invitation to all for help with this. I would love clear and concise robot rules, bumpers and everything else! Please help me help everyone!

15 Likes

yep yep yep

This doesn’t seem too crazy for me; I think the money can make sense. I want to say Robopromo(?) proposed a “D” shaped cross-section pool noodle alternative that was shot down. I also didn’t love it because it seems like it would be “easy enough” to go a step further and integrate a durable backing and outer layer. Perhaps even nix the backing, and ship a roll of the stuff you slide your cut plywood into, then staple/adhesive together. Sell adhesive numbers specifically compatible with the outer material to eliminate the classic peeling vinyl and that’s around half the mistakes cut out.

Buying a 80% bumper all from the same supplier doesn’t seem like too much of a stretch for FRC.

3 Likes

Personally, the worst parts of bumpers for me are two things:

  1. Precisely adding mounts to a squishy fabric backing such that it aligns with mounts on the robot.

  2. Sizing the bumpers appropriately relative to the robot frame in a way that isn’t too tight or too loose while also accounting for fabric thickness.

I agree that we need to change the bumper rules to make their manufacturing and integration easier for teams. I don’t know exactly how, just yet, but I will think on it and post my thoughts.

15 Likes

The heat transfer (iron on) vinyl is such a better option than the standard adhesive. It costs the same and anyone who can cut one kind can cut the other. I wish everyone used it. And also I wish everyone remembered to apply numbers before wrapping the fabric around the bumper.

4 Likes

These are the ones I find on the field carpet more often, weirdly. It seems like some fabrics don’t take them as well? Or need a hotter iron? Not an adhesive/iron-on expert, just a complainer lol.

But yeah, once the bumpers are assembled please just get some white paint and a paintbrush, or make a duct-tape stencil and rattle-can them. As hilarious as melted bumpers are, they’re pretty depressing a few moments later.

3 Likes