This year we started off having terrible experiences with our bumpers, and made it one of our must-fix things after our second competition. What we ended up with is by far the best solution we’ve ever had, and we will very likely be running it again next year.
V1!
We started off by using L brackets that we pinned to the frame, and indented corner brackets. The corner brackets were made by cutting a 3x3 1/8 wall tube in half.
The pins occasionally got stuck and would require a large amount of force to remove, and people removing them would hit their hands on whatever was above the pin. This was a safety hazard as well as slowing down our bumper changes.
They stick into the frame, requiring us to bend the bumpers around our superstructure, which went all the way up to the edge of the frame. While we never had a bumper fail while doing this, it could after a few hundred cycles.
There isn’t enough space for our corner brackets, so we had to rout out pockets in the corners to fit them. This was a pain to manufacture, and was really time consuming.
They didn’t support the bumpers enough in the z-axis, meaning we couldn’t use the bumpers to pick the robot up. We really like doing this on our 2020 robot.
The front of the bumper had no support, so it hung lower than the rest of the bumper. This caused our kicker bar to sit at the wrong height, which cause intake problems.
V2!
We had an intermediate step to solve these problems, rivnuts. These were installed in the frame rails directly, and removed the safety hazard of pulling out the pins. We added a flat spot to the bottom of a bolt to lessen the likelihood of stripping the rivnut, and never ended up stripping one.
The rivnuts were hard to install, and we didn’t have a quick way to fix them if they stripped at a competition.
The modified screws were separate parts, and got lost fairly frequently.
They didn’t solve the stability issues from V1, nor did they address the manufacturing pain of insetting the corner brackets or bending the bumpers to get them on the robot.
V3!
To finally address all these issues, we created our V2 mounting system.
This system used screws tapped into the frame rails, onto which the bumpers slid on. The bumpers were then prevented from sliding off by a screw on each side, which takes almost no load.
The mounting system does not stick inside the frame at all, allowing us to slide the bumpers on an off with ease!
The plates on the bumpers stuck the bumpers out enough to clear the corner bracket, so we didn’t have to rout out a pocket!
The rails supported the bumpers in the z-axis, and were constrained so we could easily pick up the robot by the bumpers
The length of the rails meant the front of the bumper was held solidly, so the kicker bar worked as intended.
The bolts to attack the bumpers were 1/2" 1/4-20 bolts, which we had close to 100 of sitting in our toolbox.
But there are still improvements that can be mode on this. If something besides a screw were used to hold the bumpers on, no tools would be needed to install the bumpers. If we used shorter rails, the metal parts would weigh less, but might make the bumper less rigid. With our small frame this year it wasn’t an issue, but it might be in the future.
This is a really cool approach. How many bumper changes have you done? My experience has been that rivnuts and tapped aluminum wear out over a season on frequent bumper changes, so we use captured nuts that we can swap out when they start getting worn.
@NatsirtD It looks like the spacing between the bumper sections (front and rear?) with the rails attached has to be pretty accurate and consistent for them to engage the screw heads sufficiently securely. How do you achieve this with both the red and blue sets? Do you have some assembly procedure to guarantee this? How do you account for the thickness of the bumper fabric on the inside surface of the bumpers, especially at the corners where one tends to get multiple layers?
We went through dcmps(12 matchs) and worlds(13 matches) with these bumpers, plus probably 30-40 between outreach and drive practices? Were going to 2(maybe 3) offseason events with these same bumpers, so that’ll be at least another 24, probably more. I’ll make a post once weve been to both, or when the bumpers fail.
The wood planks on the bumpers determines the engagement, and it’s fairly easy for us to nail that length.
To locate the rails on the bumpers, we measure from the bottom of the wood and the corner of the bumper, so that the front of the bumper is a good distance from the front of the robot. (Forgot to mention this in the original post, the rails are on the sides of the robot, with the entire front/back being covered by a single bumper)
To account for the fabric, the bumpers are oversized by 1/4" in width, and 1/8" in length. We staple the fabric then screw the plates into place on top of it for added security.
@wygrangea would know more about how the fabric was done, I wasn’t there when it happened.
First off, I love this idea (and might be stealing it for 2023). One question though: I’m not sure how R410 applies to this system. From what I can see, R410 states that the at least 1/2" of each wood section must be backed by the frame, with less than a 1/4" gap. Looking at the CAD, the aluminum track sections are 1/4" thick, so the gap is at least a 1/4". Or do you treat the bolt heads as the frame perimeter? Have you had any issues with this?
Didn’t have any issues with this at DCMP’s or Worlds, we’d been counting the bolt heads as part of the frame perimeter. There could be some weird rule interactions with the bolt head exemption from the frame perimeter R101 and R410, so don’t go a full 30x30 and do this.
It also seems a little unclear if a series of bolt heads counts as supporting the whole 1/2" of wood. Good to hear that you didn’t have any issues with it though.
For the bumper corners we used the technique shown in this thread Bumper fabric & How to Make Your own Bumper Numbers with great success. By doing the two step folding and stapling into the top of the plywood for the under layer and then into the back of the plywood it keeps things neat. Since we used a Cricut and heat transfer vinyl for the numbers and keep an iron in the shop for use on the bumper fabric, we also used the steam function to soften the fabric and help make it more pliable. We also used straight pins to pin the fabric into the pool noodles and thanks to diversity of our team had an intern fashion designer to do the tuck folds and make things very even. The bumpers are still holding very well, though the numbers over the intake have suffered from the cargo. Definitely the best set of bumpers we’ve had in recent history. Plan for our two off-seasons is to paint in the number that has rubbed off since it’s hard to heat set vinyl onto fabric on pool noodles and see how that holds over the matches. Having personally changed the bumpers in the last month they are holding up just fine, though we have to jiggle them more than we used to to get them off, but they still line up and attach just fine.
We had 1/4 20 riv nuts this year going through 2 regionals plus champs and they are fine. We had split bumpers so 4 bolts per side. It took us around a minute or two to swap bumpers using 2 drills.
