The Everybot crew is having a great time unveiling the FTC Everybot on FIRST Canada LIVE!
Below is the 2024-2025 FTC Everybot Reveal Video and the currently available resources.
That Onshape document is a work of art. Flip through it even if you donât use Onshapeâall the configuration options (REV/AndyMark/Tetrix/GoBilda motors, imperial and metric PVC, multiple servo options) scream incredible attention to detail.
Canât wait to see a few of these take the field!
All credit goes to @yapple.
The 118 students did a great job designing the parts but we certainly didnât have them organized in the best way for the public release of this project. Learning moment and we will be better next year!
Jaclyn did a great job taking our mess and making it one of the most intuitive and incredible onshape models I have ever seen.
Wow. That was not what I was expecting. I had expected to see a traditional side panel chassis frame. As a team that extensively uses 3D printed parts and extruded aluminum in our own designs, hatâs off. That tape measure drive gives me warm memories from Relic Recovery and Rover Ruckus.
We will play around more and offer additional feedback later, but one thing struck me right away, so many unrestrained nuts. In future, please look at building in captured nuts or better yet heat set inserts. Weâve moved to heat set inserts for our 3D printed parts almost exclusively and the skill set for using them is easy for students to pick up and that saves you needing to put a wrench on all of those nuts.
We intentionally did not do nut pockets or any sorts of inserted thread types such as heatset inserts because we support M3, M4, 6-32, & 8-32 thread sizes. Each with a different sized hex nut and hole required for heatset inserts.
This would be a fantastic way for teams to improve the design on their own however!
Great design - I love the functionality and build system flexibility. I do have some concerns about the directly-driven wheels, however. Directly-driven wheels without external support increase the static and shock loads on the motor gearboxes, causing them to break much more frequently (especially with spur gearboxes). To mitigate this, it would be best to incorporate a bearing support on the opposite side of the wheel or to use a separate drive shaft.
From a strategic perspective itâs great; it has a mecanum chassis (albeit a kinda slow 4 ft/s in the rev mec config), a good intake and the ability to score in high bin. Seems pretty consistent with the overall goals and mec makes it (in my opinion) a bit easier to write autonomous routines for.
Vendor kit-bots fall into this weird category where they want to demonstrate what their starter kits can do but donât want to tell teams âhey go buy this $500 drivetrainâ so they compromise and build a 2-motor tank drive few teams actually build in practice. Fitting in a mecanum chassis here is a genuine achievement and you should feel proud.
That said, I do think it would be a value-add (especially this late in) if there were cad configurations showing how you might attach the scoring mechanism to some of those kit mecanum drivetrains though. Newer teams in particular will likely already have them; while thereâs some selection bias here, if you watch various early meets from all over thereâs a ton of teams that build a rev or gobilda mec chassis first and then worry about scoring mechanisms second.
While on paper the bom cost ends up more expensive, in practice itâs subsidized by the team already having the chassis.
Me personally, if I were to build this thing, I would probably want to use a kit chassis with it; from a strategic perspective theyâre known and proven solutions and if the following season I wanted to build something different itâs a chassis I can reuse in a different game.
I am also kinda curious to see how this design holds up over a whole season as well. There are some interesting design techniques used here and it would be informative to see how they hold up and what sort of maintenance they would need over time. Time will only tell, however.
Weâve posted these resources in many different locations, and these are some of the most seen questions so far across every shared platform. I wanted to post them here so that we had a lot of the information all in one place.
Thatâs a lot of printing⌠Are you sure?
Yes. Positively sure. As mentioned before, we used the Microcenter $70 Ender 3 with Inland PLA+ and the parts printed and worked just fine. The parts off of that printer actually worked better than the PETG ones printed on our Bambu X1Cs.
.
One of FIRSTâs requests to us was to heavily rely on 3D printing. We obliged and we hope that you can learn a lot about 3D printing by comparing your designs to ours. I have personally been designing parts for 3D printing for over a decade now. I passed along everything I knew to our student designers and they implemented it all amazingly.
.
Additionally, I personally bought an Ender3 V3 SE and printed one calibration cube to check to see if the printer could actually print parts. We did zero tuning to this printer and the second print it ever printed was the large herringbone gear portion of the arm. It printed successfully on the first try. After this test I donated it to one of the student designers who did not have a printer of their own. This student then proceeded to print most of the prototype parts on this printer. I also test printed parts on my personal Prusa Mk3S+ and super old CR-10. All parts on all of the printers printed successfully without support.
.
My old teams used many 3D printed parts when I was still mentoring FTC before 2020. We also use many 3D printed parts on our FRC robot so we know that parts like this can withstand a bit of a beating.
But it is also best to be prepared and print spares of anything you are worried about breaking. We recommend
How does the tape measure work? Why did you choose this mechanism?
The tape measure extension works by using two compliant wheels to press into the tape and when they spin, the tape measure is pulled out of the tape measure casing. We decided to not modify the tape measure in any way so that we could support any range of tape measures as long as they fit in the slot in the arm.
The main goal of Everybot is to be a simple and accessible robot. We started off our design discussions with the standard cascade and continuous elevator type designs and decided that there are many drawbacks to them when used in this orientation.
.
First being that they are heavy, and part dense. Second being that they are much more complicated with multiple string, chain, or belting routes around pulleys or sprockets.
Second, they require much more control. You need some sort of code feedback to not overdrive a standard extension mechanism.
.
One thing that the tape measure extension does that no other extension mechanism does with such grace is showcased in the reveal video at 0:15 is that the extension can actually protect itself from damage. The tape measure will retract when the arm is forced into the field perimeter, saving the robot from damage. (This is a perfect time to mention the biggest reveal video flex Iâve ever seen from 6800âs 2022 robot )
.
This is also true with the control. The motor driving the tape measure just needs to run either in or out and the extension will move either in or out. Once the limit of extension or retraction is reached, the wheels will act like a clutch and slip on the shaft or the tape and no damage will occur from over driving the mechanism. You can see this in the reveal video at 0:05. Pay attention to how the wheels keep moving and the tape measure stops at about the 6â mark and just past the 18â mark.
The intake looks like it should just be a claw. Why shouldnât I just use a claw?
This is a good question, we decided on using this type of intake for a lot of reasons, but I will highlight a few here.
.
We found that a claw, by nature, needs to reach around the outside of the Samples. This is all well and good until there are multiple samples next to each other and you want one in the middle of the pile. The claw cannot grab a sample that is not already singulated. With the side roller design, it is able to kick away anything that is not centered. You can see an example of this in the reveal video at 0:14.
.
This intake design allows the Samples to be misaligned. Unlock some other intake designs we tested, top roller and claws specifically, this design allows for the most misalignment, giving teams the largest acceptable intake envelope. You can see a great example of the Sample alignment at 0:44 in the video.
.
This intake also auto-orients the Sample. We decided this was important because scoring the Specimen âlong-waysâ allows for more densely packed Specimen on the Chamber. You can see how closely the robot can score Specimen together at 0:32 in the reveal.
Are the motors really only held on by two screws?
Yes, but no. Every motor on the robot is actually secured by zip ties around the entire motor body. The âmountingâ screws really just act like pins. The bolts are only necessary to stop the entire motor from rotating within the housing.
.
We spent a lot of time testing this design and had zero failures from anything a robot will see during a match. We do caution teams that modify the robot to climb to the top Rung about falling and landing on the wheels. These were not designed to withstand large drops like this. This was the only failure we observed.
The robot scores in the high Basket?
Yes! Even though it was not on our Everybot Will list, we found that we could gain that functionality pretty easily if we put the arm pivot at the right spot. So we did. We hope this makes the robot fit into most alliances that much easier.
Why didnât you put the robot on a supplier specific chassis?
Supplier specific is sort of the opposite of what we aimed to do in this project. We did however think of this. And that leads to the decision that everything above the drivetrain is supported by two vertical uprights. We simplified the uprights to create a superstructure that could easily be adapted to any pre-existing chassis.
Is the video really 1:19 and not 1:18??? Missed opportunity!!
Yes⌠YouTube does this really fun thing where videos are 1 second longer on mobile than it is on desktop. @Jennakayhas already given a much better answer here.
.
We chose 1:18/1:19 rather than 1:17/1:18 for the Everybot because during the FRC Build season we have quick update sessions at the full team meeting. We do 1:18 of either FRC Robot or just general âgood stuffâ and then we do 1:19 of Everybot. Just because we believe Everybot is that important, it gets more time.
Michigan teams are not being served as well as they could have if this robot was released earlier. Why was it released so late?
Yes, itâs true that releasing this earlier would have been much more beneficial for Michigan FTC teams. FIRST just asked to start this project, and this is the first year of this initiative. We made the decision to spend extra time to release a much better design with a lot more time and thought put into it. We figured that releasing a bad set of documentation sooner would actually have hurt teams more. Because we have now laid the groundwork, we believe next year will have a much sooner reveal date. Hopefully no more than 4 weeks after the game is unveiled.
.
On a more personal side of things, we discovered there are many very important things that happen in September for High School Seniors. SAT Testing, College Applications, Homecoming, etc. We overlooked these in our initial planning as we did not know what we did not know. We will be better in the future.
Is 118 doing this just for Impact??
This was brought up a lot and it feels very weird that this is the reaction to this robot release. No. We donât do anything strictly for Impact. As a team we have decided that if we are doing something purely for an award, we arenât being genuine and therefore doing it for the wrong reasons. If we were just doing this for Impact, we could have spent the last 2 months doing way more outreach events rather than spending hundreds of hours on a single robot and document release package. We do what we do because we are passionate about it. I am personally more passionate about the Everybot than I ever will be about any 118 robot. I know other members of the team feel similarly.
I hope this helps!
If there are any more questions, comments, criticisms, or concerns, please do not hesitate to put them here.
Quick note for anyone thinking about saving weight by taking apart the tape measure, please DO NOT. I had a broken tape measure in 10th grade and I thought it would be cool to see what was inside them. Turns out it is a giant and very powerful constant force spring. Luckily was not hurt. It exploded when I opened the case and broke apart into a few pieces. Please DO NOT DO. I would greatly not recommend.
On the opposite end of the spectrum, my uni club had one that wouldnt work smoothly and i took it apart to fix it. It did get fixed but in the process of trying to tame the cf spring so i didnt have to rewind it later, i broke a plastic part of the shell that slightly affected the locking mechanism.
I still disagree. I think having configurations for common kit drivetrains is vendor-agnostic, just at a granularity more reflective of FTC team purchasing decisions.
While mecanum has the properties of swerve from a competitive standpoint in FTC, the Rev/goBILDA/AndyMark/Studica/etc vendor chassis also has properties similar to the AndyMark AM14U kitbot chassis in that they are well-understood, perennially reusable, guarantee that your robot drives the entire season, and something that thousands of teams have lying around in some form.
Again, the vast majority of new teams seem to buy either a Rev or goBILDA mecanum drivetrain kit and just run with it for the next eternity. These kit chasses can have pretty long shelf lives; there are teams out there sporting Strafers from 2019 which would almost certainly predate all the students currently in the program. While on paper these drivetrains can cost a bit more than custom drives, they become incredibly reusable investments that many teams hold onto for the rest of their life in the program.
Thatâs not to say building a custom drive at this scale canât be a good option, but it isnât necessarily representative of the median entry-level teamâs needs here. The current solution could be useful for ancestral teams that never bought a kit mecanum chassis but accumulated the parts for mecanum drives over the years and cobble something together, but teams founded after 2019 tend to not to be in this scenario as they will have most likely bought a kit chassis instead.
I guess to put it in FRC terms though, it vibes like if the FRC version of this project rejected the AM14U chassis for being too vendor-specific, and then developed a custom WCD with CAD configuration options for AndyMark Toughboxes and WCP gearboxes, and then saying at the end that the robot is pretty easy to adapt to an AM14U chassis anyway.
Like sure, it may be theoretically cheaper and fully interchangeable to what teams may have lying around, but I donât know how well it fits the program overall.
(I know itâs not a 1:1 comparison, but itâs the closest approximation I could come up with)
The community will cover this very quickly. My team spent the weekend helping our FTC team build the mechanisms of this robot and integrate it onto the gobilda staffer chassis, one of the more common options.
It wasnât difficult at all. Adapting it to rev (and probably the Andy mark chassis) will be just as easy. Tetrix might be tougher, unfortunately a lot of the teams that need a design like this still use tetrix. We worked hard to purge it from Detroit schools a couple of years ago, most of it is thankfully gone.
If it were my choice to make I think releasing plans and instructions for common drive trains would have been helpful. However thatâs also a lot more work, itâs not just a matters of releasing the CAD. They have to test the configurations, write the documents, and support those use cases online.
And the integration wasnât bad. The design leads itself well to the standard chassisâ with some very small changes. Weâre going to be supporting more teams in this regard so weâll publish our findings.