Electric Mayhem White 12737 2024-25 Build Thread

[njj2026]

Welcome to the build thread for 12737 Electric Mayhem White’s Into The Deep season!

General Info:
For some general information, our team is primarily focused on the development of our members, being the traditional “jv” team, while our sister team, 12736 Electric Mayhem Green, is the traditional “varsity team”. We have 17 members, some with more experience than others, but our goal is to, by the end of the season, have the members of our team be at the level where they can move to our sister team
Our team started in 2017 as a school based team in Buffalo, NY, with our first game being Relic Recovery in 2018.
This team specifically is one of our school’s 6 FIRST teams: 3 FLL teams made of middle schoolers, 2 FTC teams with a combination of middle and high schoolers, and 1 FRC team built out of only highschoolers

FTC Open Alliance:
As members of FTC Open Alliance, weekly updates about our team will be posted to this build thread.

[njj2026]

Last weekend, we as a team participated in the 30 hour challenge. This was hosted by Green team, but many of our members came along, and in total, the challenge was a success for us.
The following update will discuss primarily our strategy and design

Strategy:
As a group, both Green and White team, after looking at the qualification rules, agreed that we should try our best to compliment each other, as this year, the first pick of the winning alliance qualifies alongside the winning captain and 1st place inspire. As the early brainstorming started, we converged on two main strategies, either focusing on sample scoring or specimen scoring. However, after some consideration, we decided that it would not be too difficult of an ask to try to do both tasks

Design:
After the challenge, this was the final version of our CAD:

Diving into the drivetrain first (pun intended), we decided that we would use a kiwi drivetrain. We decided this because it would keep all the wheels on the ground, and it gives us an extra available motor slot to use elsewhere.

For the manipulator, our inital design was to use a GoBilda viper slide with a 3D print attached to it. This was so that we could score both samples, as we could use the full extended length of the viper slides, and specimen, as we could simply not extend all the way.

And the last major part of the design, the climbers. While not fully imagined in CAD, our initial idea was to use a combination of two viper slides to ladder climb up to level 3. The reaspm we need to ladder climb is because the rules specifically state that if you plan on doing a level 3 ascent, then you cannot be touching the ground while you are touching the top beam. We were basing our design primarily off of Green team’s prototyping, as it would give us more time to prototype our main robot.

That’s it for the 30 hour update. Expect weekly updates coming from myself and the co-captain Austin.

[DjDog56]

9/14/2024, One week into the FTC season
Strategy:
Both us and our sister team decided on having complementing strategies, so Green team is doing a specimen strategy while we do a sample based strategy. We are planning on scoring 3 samples in auto and then cycle the high basket during tele OP and then doing a level 3 climb in end game.

Design:
Our drive train is going to stay the same as the one we thought of during 30 hours, which is a kiwi drive train.

For the manipulator, we are making big changes from 30 hours. We are now going to do a internal hand off. We will have an extending arm to grab samples from the submersion zone, that will deliver samples to a passive bucket that will dump samples into the basket. We are experimenting with REV T-slots because viper slides are heavy and slow. With REV T-slots we are hoping for faster extend and retract speeds.

With the climbers on the robot we are still with our initial idea of 2 viper slides because of the strength of the slides with them being able to reliably lift ourselves up.

[njj2026]

Week 2 after kickoff:
Strategy:
So far, after last week’s discovery with the internal hand off for sample scoring, our strategy has not changed.

Design:
This week, we primarily focused on skill building and development for our newer members, but on the side, we begun work on our foam model.

Our plan for the foam models is a 2 sided strategy. First, it gives us an opportunity to check if our geometry is viable, and if we can fit in the 18x18x18 sizing cube, and second, it allows us to easily divide the CAD work among our members, as we can simply write on the foam who is going to make what

[DjDog56]

Design:
This week, we focused on finishing our foam model as we are waiting for parts to come in from the mail.

We will have 2 slide systems. The first set is to go into the submersible and retrieve samples. The second set will go up and drop the samples into the bucket along with climbing. We are still experimenting with how to do the ladder climb.

[njj2026]

Week 5 update:

Design:

Big design updates this week.

First of all, we finalized our design for our robot. For the climb, which was a point of difficulty for us, we plan on doing a similar first half to our sister team. We climb up to the first rung, then latch on with another set of hooks. But the second rung is where we divert. We plan on reviving an older climb design from last year. In the design below, we hook onto the second run by extending after we reach the first climb, then hooking on with the hooks you see, and then pulling a winch to have the hooks most likely coming of magnets or velcro in order to fully reach the level 3 ascent

More, we have begun work on the CAD, with the basic drivetrain being done, that being a kiwi plate with motors and spacers. Its basic, but it is simply a starting point for when we fully distribute work.

More updates to come

[DjDog56]

Week 6 update

Design:

This week we have made massive advancements to our CAD from last week. We were able to turn a majority of our foam model into CAD.

This is all the progress we have made in the CAD this week. We first were able to get the full drivetrain into the CAD model. Then we were able to put our electronics into the CAD by putting them on the right side of the upper cookie. We also were able to finish our passive dumper that we will be using to deliver out samples into the baskets.

Along with this we also made changes to our intake. The biggest change we made is that we are using a Swift Slide instead of a Viper Slide. This change should make our intake faster. Along with this we are also going to have the hand on our intake rotate so that it can deliver the sample into the outtake

The hand will be connected to the swift slide through a L bracket and there will be a servo that is mounted on that L bracket. Then the rest of the hand will be connected to that servo, so that it can rotate.

[njj2026]

Week 7:

Massive update this time. We have made massive strides in the CAD, as well as changing our strategy so we can have time to prepare for our first competition

Strategy:

This week, we discussed it over, and decided that we are going to abandon our level 3 climb, and instead focus on more cycles during endgame, which can theoretically gain us more points. If we can score even two more samples during endgame as well as gain even a level 2 climb, we can get 15 + 2 * 8 = 31 points , which is 1 point more than if we did a level 3 climb. The big motivation for this change was so that we can have time to build, program, and especially have time for drive practice, so that we can be ready for our first competition at Penfield.

Design:

Now, CAD time. This week we have essentially finished the entire CAD. The only things that we need to finish is putting an encoder on our GoBilda Viper slides, and that is a discussion that we will have during our next meeting

But this is our current CAD:

And from the back:

So the big changes. Starting with the intake, we mounted the swyft slide to the bot, and we modeled our hand based off of team 10355’s hand from their Robot in 3 Days design.

This is mounted to a servo on the slides, as shown below:

This hand then turns 180° to face our outtake, which goes up on our climbing slides to reach the basket

And as you can also see, there is a spot for specimen intake and outtake, which will use our new compliant filament to snap into place around the specimen

Finally, the climbing has also been tested and works in CAD, as shown here:

(This is a previous version, but the hook design has not changed since)

This week we plan on finishing CAD for good, and hopefully begin the manufacturing process, which can help teach new parts of robotics to people who have not done this before

[DjDog56]

Week 8:

Strategy: We decided to go back to a level 3 climb, because of a rule clarification where you are able to brace yourself up against the submersible while going for a level 3 climb. With this clarification it would make the level 3 climb a lot easier to achieve so we are going to attempt it.

Design: With the CAD, we have added secondary climbers for the level 3 climb. Along with this we have also added all of the sensors into the CAD that the programmers wanted. We added a color sensor, NAVX, laser light odometry, and LimeLight for position data. Absolute encoders on the outtake slides, and on the intake slide for go to position data. A color sensor on the hand to tell if we have the correct block or not.

This is the full CAD as of right now. With the first iteration of the secondary climbers and all of the sensors on it.

Programming: This week was one of the first big week’s programming wise. We have planned out all of the methods that we want to have and a basic auto design. We have also started programming some of these methods so we will be able to push code and start testing as soon as the robot is built.

[njj2026]

Week 9 + 10:

Apologies for the late build thread, so this will include the last 2 weeks.

Week 9:

First and foremost, this week was a lot of FUN.

With that shameless (and a bit late) plug out of the way…

Design:

This week was our final CAD push, and we were able to get the CAD done in time for our deadline, halloween.

Halloween version of the CAD

But most importantly,

Routed bottom plate

We have started assembly

Week 10:

This week was a major push for the assembly. This weekend, both green team and ourselves came in to make major strides, and at the end of the weekend, this is what we have came out with

The version of the robot that came out of the build weekend

It was by no means complete, and some geometry issues needed to be addressed, specifically with the climbers, so those got re-CAD-ed into these

Additionally, if you would like to watch us work, both live streams are up on the Electric Mayhem Robotics youtube page.

This being said, the week went on, and the programming team was able to get what was available with the robot to drive. In the following weeks, we should be able to, surprisingly, finish the robot and get drive practice in for our first competition in about a month at the Penfield qualifier. So, this most recent weekend, a few of us came in and got the robot nearly completed.

The most current physical version of the bot

Now looking at the electronics, similar to last year, we are using a NavX micro for our IMU, and then for odometry, we are going to use the SparkFun Optical Odometry, which should save us a lot of physical space on the robot, as we no longer need to worry about dead-wheel odometry placement. Both can be seen here

The odometry sensor is under the NavX but it is there

Additionally, we decided that unfortunately, due to our build geometry, we will be abandoning the level 3 climb once more, as we fear that we cannot successfully make it, and that we would be better off going for the extra 2 cycles

With that, adios

[njj2026]

@0xCUBE
The link should show the most recent version now. There are multiple folders in the document, so you may have just been seeing our 30 hour folder

Hope this helps

[njj2026]

Here ye here ye

I come to announce that team 12737 Electric Mayhem White is in fact not dead, but very bad at making “weekly” updates

With about 10 days until our first comp at FTC Finger Lakes Qualifiers in Penfield, an update is very likely due

Strategy:

You guessed it, we have gone back and scrapped our 30 pt climb in place for a few more cycles. This was mainly due to an oversight while designing our geometry, and we simply could not find a good place to successfully pull this off. Doing calculations in our lab, if both our sister team and us can accomplish our theoretical goal, we can get a score of over 350, beating the current world record. And that would be at our first competition. Needless to say, if that happens that would be amazing

Design (In the most literal sense)

Yesterday, one of our members was able to finish the design on our skirt, and in my humble opinion, it turned out fantastic, so after that day, this is our current design for the bot

(most current version of our bot, with the vinyled skirt)

Practice for Penfield

With our bot being this close to being finished, enough code has been added to the bot that we can begin drive practice very soon (we plan for this weekend). On monday, while testing some code, we got a full video of a full cycle, which can be seen here

cycle times

With that, adios, and wish us luck for our first comp

[njj2026]

Blockquote
https://youtu.be/JKhNxD_-zIU

ignore the flying block at the end, that was because I ran way too fast into the basket

[Rocky_S]

saw yall in FLR last year, keep up the amazing work!

[guineawheek]

@njj2026

Now looking at the electronics, similar to last year, we are using a NavX micro for our IMU, and then for odometry, we are going to use the SparkFun Optical Odometry, which should save us a lot of physical space on the robot, as we no longer need to worry about dead-wheel odometry placement.

Given that the SparkFun OTOS also estimates heading, what’s the application of the navX-Micro here for?

[njj2026]

This is our first year with the OTOS, and we are not entirely sure how it works, but we do know how to use the navX imu. our sister team is doing experiments with the OTOS so if they see that it is better then we will switch

[njj2026]

R and R, my favorite activity.

Reflections and Revisions

Penfield

So yeah, penfield was yesterday, and yeah it didn’t go great per se.

(This was me after our DS crashed for the upteenth time)

However this is the joy of robotics. We as white team still plan on doing two more comps, one in our home town of Buffalo, and then the one out in Utica. But the event wasn’t a total failure either. First and foremost, if any of you tuned into the livestream of the event, the emcee did call me out by name for being a good coach, but the other big thing that happened,

We won the Innovate Award.

In my opinion, this was a bit weird as they talked about two things that weren’t really ours. Those two being Mario Kart (RoboKart is the actual name Nintendo don’t sue us), and 30 Hours, both of those were championed by green team. However talking with other people really set in stone that this was not the worst outcome possible. Mainly because both of these two things drive home what this whole entire team was meant to do at the beginning of the year. Be a learning team. The Mario robots were extremely useful as in the beginning of the season, we took them apart to help green, but in doing so, we were able to teach a lot of the members about how these robots work. For the mechanical subteam it was good to learn how to kill a bot, and the programmers got to program the new versions with some help. And a similar thing goes for 30 Hours; it was a great learning experience for everyone on the team.

So yeah. While we didn’t qualify or do too well in the robot category, we still did do good, and we have a lot to learn for the next comp.

Speaking of…

Revisions

So what went wrong

First and foremost, our DS crashed on us multiple times. As in, every single match we ran an auto the screen would turn off and the bot would disconnect, which really did not help our chances. Talking to people about it, a volunteer at the event gave a really good possible reason as this could be a grounding issue, as the robot’s building charge as it drives and it eventually just crashes. However the reason I think is more plausible is a reason given by a friend of mine on green. The issue could be that, because we are using external power banks to power our DSs, they every once in a while can turn off and on really quickly, which is just enough time to completely kill our bot. This will definitely be solved by just using the DS batteries instead of a power bank.

So now how do we fix all of these issues?

On the bus back home, a few other people and I came up with a Google Doc with stuff that needs to be changed.

Starting with programming, we have a limelight on our bot, however we did not have any time for this comp to get it working, so it was essentially a deadweight. This will be changed as we can get more reliable autos if we have localization on the field. Speaking of autos, last year, green team made a revolutionary new auto called spline, which, as you can guess, moves the bot in a curved path. I won’t go into spline as it is not ours, but the reason we couldn’t get it to work on our bot was because the way it was written, it required code to be made a different way than it was already written, So for next comp, the entire codebase is going to be purged so that we can use spline for better autos. And finally, if we do a purge on the codebase, we will also be updating all the firmware on the hubs and DSs so they can be more reliable.

Moving on to mechanical, redundancy is a massive thing we need to work on for next comp. At the very start of the day, our wrist broke, and we duct taped the hell out of it as if it became broken, we would be decommissioned for the entire day. So we need to build redundancy for these parts, and if we can’t, print spares of everything so we don’t run into this issue next time. The next point here is slides. Our intake is only driven on one SWYFT slide, and because of that, it sags a lot, making it a bit scary to use quickly, as it can run into parts of the bot it probably shouldn’t run into. Following that, the intake is going to be improved as we need more reliable intaking, as it currently can move blocks into orientations that are very difficult to grab, so we plan on possibly adding flanges to the sides so we can get the samples in easier. Finally we have an idea to build protection for parts that are meant to break, so if the protection breaks, the parts that aren’t supposed to don’t.

Finally, on the pit next to us, we saw team 12833 Mechanical Meltdown use different types of batteries by GoBilda that are different dimensions, and apparently are just overall better than the REV ones. Additionally, we plan on just buying new DSs because of the crashing reasons I explained earlier.

So that was penfield. See us in Buffalo soon. adios

[njj2026]

quick schedule correction, we are going to MVCC qualifier next, then buffalo

[njj2026]

“From the earth I rise, and to the earth I will one day return”

~Colonel Cornelius Cornwall

We’re so back

As you can see, we’ve been hard at work. We added the majority of changes to the robot, with the GoBilda Batteries, and our own way of keeping them secure, adding a second slide for stability, and hopefully working autos.

(the new shell for the GoBilda batteries)

(carbon fiber 3D prints for extra reliability on the wrist)

(both slides in action)

Additionally, since we added another completely passive slide, it weighed more, so we decided to use 312 rpm GoBilda Yellow Jacket motors, as opposed to 1150 rpm. This should also help with torque issues and we should be able to hold the slides in place and be able to reliably intake and handoff

Outside of that, for comps, we are learning from our mistakes and are going to print extras of every single piece, so that if a piece breaks, we don’t need to bandage it up with duct tape. And since our wiring was questionable last build, on our cart, we taped a QR code that has a link to a wiring config.

Programming wise, tonight both teams will be doing a video call to do group programming so that our spline issues can be solved so we can fix issues like the ones green had during their Penfield experience

12736: Electric Mayhem Green, 2024-25 Build Thread

SUNDAY SUNDAY SUNDAY! FIRST TECH CHALLENGE FINGER LAKES QUALIFIER! Well Well Well, looks like our first competition is basically here, like tomorrow. New Intake We didn’t like our previous intake. So we changed it, going back to our old ways of over-sample intake. The new intake has a very wide intake range and is much simpler. This intake has worked great, and is exactly what we needed in an intake, fast and versatile. [image] New and Improved Over the Sample Intake Upon testing of our p…

After break

Once we get back from break, we will be doing as much drive practice as possible so that we can be extra assured that when we head to Utica, we will be in good shape. Additionally, our skirt will be covered in a clear vinyl coat so that it is more protected against battle scars and what not

With that, hope everyone has a happy holiday and see you in the new year

[njj2026]

The public cad should be updated on our FTCOA page