FRC 5414 Pearadox 2024 Build Thread

[ahartnet]

EDITORS NOTE: Posting this on behalf of some students, but I’m excited to share the Pearadox will continue a build blog in at least some capacity this year. I’ll try to clearly state when posts are coming from me or when I’m posting on behalf of a student.

Annual Round Robin #X

September 27th 2023

(another Editor’s note: I’m not sure if X is because they weren’t sure what number this is or the roman numeral X because it’s our 10th season, so I’m leaving it!)

On September 16th, we hosted the annual Round Robin training for rookie members. The Round Robin training is an annual training for new members where they participate in and learn from engaging activities and presentations at multiple stations that can change year to year. This year’s stations were:

• Programming and Electrical - How to use software we use in robotics and learning the basics of the electronics in a robot.
• Prototyping, 3D Printing - Essentials to designing and creating a robots.
• Machine Shop - Learning how to safely use drill press, CNC Router, lathe, and other power tools.
• Components, Design, and Onshape - Learning how to CAD design, and learning robot components.
• Impact - Learn about marketing, community outreach, and media for Pearadox
• Coach’s corner - learning about what makes Pearadox Pearadox, why the mentors do what we do, and answering any questions folks might have. Coach does this to try to be less intimidating to our new students

This year’s rookies got the opportunity to get some hands-on work as they made recreational prototypes from old game pieces and snipped wires at the electrical station. We also talked about our team’s outreach and inclusivity initiatives we’ve implemented, such as Brilliant Black Builders, Gearbox Girls, and Safe Spaces in STEM. Veterans were treated to an advanced Onshape workshop led by Spectrum FRC 3847 Head Mentor, Allen. To recap, this year’s round robin ended pretty well helping our new members get adjusted to Pearadox and helping them prepare for our competitions. In fact, one of our first competitions this Fall was Remix, which we competed in last weekend!

Special Thanks to: Veterans who did Round Robin presentations for the new members.
Credits: Zayan, Andy, Sofia.

[MisterGrapfruit]

Our team is in the process of building out our super pit, and after reading that deploying 4414’s upper case section is a 5 person job, we’ve been looking into the option of adding gas springs to help with lifting the top portion of the case. (We chose to go with the same Omega case design as 4414, so the top section of shelving folds down for transport.)

We think this might be the most promising geometry so far, it uses a a pair of gas struts on each end to support the weight as it is lifted, but it requires the addition of a triangular support plate somewhere on the top section of the case. Has any other team done something similar with their pit?

Full CAD available here: Onshape

[ahartnet]

I imagine it’s too late for anyone to use this year, but I wanted to share a schedule generator I made. It’s useful for generating a one pager of the schedule, but I mostly use it for knowing what teacher is supporting what meetings:

Copy of Schedule Generator

Some key features:

• Set a default weekly schedule, with the ability to have some “special” meetings inserted
• calculates total hours each teacher supports (makes some assumptions for competitions) and number of days teacher supports, this way we can try to keep things evenly distributed
• currently set up for 4 teachers. I imagine we’re fortunate to have more teachers than most so you could just leave teacher spaces blank
• highlights meetings that don’t have a teacher assigned
• generates a list of days a particular teacher supports

It is not as robust as I would like - one day I hope to make it more adaptable - there’s lots of things I have more or less hard coded for Pearadox’s set up. But it could potentially be a starting point for others.

[ahartnet]

Just a note, I got to use this last night and there’s some bug where some of our competitions got assigned all to the same teacher in the generated schedule tab, despite being assigned correctly in the backend data. That’ll be a job to fix later. But going and overriding the teachers from the drop down in the generated schedule did seem to produce an accurate summary table.

I think that’s the only known issue at the moment that I ran into

[ahartnet]

Going to split into two posts. Below is a post from some of our great underclassmen that are wanting to take some of this on. I’ll double post for some other thoughts/learnings I had from some testing today.

I also did not realize that we were posting directly to our blog page on the website. Check it out here for some posts about some outreach event (Trick or Treat Trail, some things we’re doing for all our FTC/FLL teams, and our princess with power tools event): Pearadox - Blog

==================

January 6, 2024
2024 Kickoff meeting
On January 6, 2024, Pearadox hosted the annual kickoff event. During kick off this years game animation is released simultaneously around the world for all FRC teams to view. The purpose of this meeting was to officially begin our season, and to create an opportunity for teams to engage in communication and collaboration with each other. This event is a great example of Pearadox’s wider outreach, with teams including Platypirates (FRC 9181) and Robo-Cults (FRC 9996) in attendance and collaboration. For a bit more detail, we have an interview with Pearadox rookie MG.

ZQ: Hi MG, how was the kickoff meeting and what did you guys do there?

“I think the kickoff went great! We layed out some groundwork for the incoming game, worked out strategies, that sort of thing. We didn’t just work on technical and strategy either, we also laid out what we were planning to do for impact this year.”

ZQ: Do you know what your impact team is going to be doing this year?

“We’re going to continue our previous initiatives, writing essays for impact awards and design shirts for our 10th year anniversary and the new competition.”

ZQ: What did you think of this year’s game, Crescendo?

“I think it was a pretty cool game. It’s simple yet complex and it’s more complicated than last years, but still not much of a challenge for us. The hoop pieces also allow for some interesting “launch” mechanisms.”

ZQ: MG this is your first year at Pearadox, what do you think of it so far?

“So far it’s been pretty great so far, I’ve really enjoyed the impact aspect of robotics and collaborating with other teams and my teammates as well, not even mentioning the events I’ve been to as well, like trick or treat trail or even kickoff too!”

ZQ: What do you think about the Pearadox blog?

“I think it’s really cool and everyone should check it out! (and like the posts!)”

Credits:
Writers: MG, ZQ

[ahartnet]

Today, we had a couple hours dedicated just for folks wanting to be on drive team. We had about 12 students show up, and a couple of others that expressed interest but had a conflict.

We spent about an hour doing a mix of an ice breaker (zip zap zop…having two different circles going on right next to each other was pretty fun of having to zone in to only your team) and talking about what I look for in drive team.

The next hour or so was spent doing some drive practice. We’re very fortunate to have a full field space and some awesome mentors, alumni, parents, alumni parents come in to build field elements throughout the week. We only have one stage set up right now (the picture below is a little old)

Our set up was:

• our tall, 2023 robot as the “offense” robot (blue alliance)
• a swerve base as the “defense” robot (red alliance)
• a stage set up on the the blue alliance side only
• do 2 cycles to the amp and 2 cycles to the point at the wing line and one of the stage legs. Repeat as many times as you can in 2:15
• no defense at in the blue wing for this week
• assume you can’t go under the stage, assume there’s a red alliance robot ready to collect the foul at the podium

Some lessons learned:

• drive paths are really limited by tall robots. Because the podium is a safe spot, a tall robot really only has one good, foul-free exit path from the source
• Getting aligned to score in the amp I think is going to take some people by surprise if they can’t get aligned with the April tags or similar. It seemed way harder for some of my drivers to feel comfortable driving sideways into it.
• as drive coach, and for my shorter drivers, it is harder than I expected to see the subwoofer. One of my shorter drivers ran into it a couple of times in the ~8 cycles they did.

All of the driving was from alliance station 2 with a wooden stage blocking visibility more than the real stage would

[MisterGrapfruit]

Alright, we’re a bit overdue for a blog post. Here’s a rough recap of our strategy, our testing with each subsystem, and our current plan moving forward. I’ll try to keep it focused on stuff that other teams might find useful

Strategy

We hope to have a sub-10-second cycle time to score 2 notes per amplification period. We intend to score in the trap after we climb, but we won’t normally choose to try for multiple traps. We expect any hoarding strategies to get shut down by opponents stealing your stash, as it adds little time to their cycle and makes you waste an entire cycle getting another note

Intake
We tried several intake configurations, most of which have already been covered by RI3D teams or open alliance blogs. Basically everything we’ve done intaking-wise has seemed to work pretty well.

In terms of geometry, we’re planning to use an over-the-bumper “assist” roller to feed our under-the-bumper primary intake. This will hopefully allow us to snatch game pieces during auton and reach into corners, but if it gets damaged we’ll still have our protected UTB intake. We’ll be doing a polycarbonate four-bar linkage for deployment, so that it can flex/retract upon impact.

Our UTB intake will be a pair of vertical wheels to center the piece and maximize our intaking width. Our OTB assist will help force the note into the wheels for the centering to happen.

Further back, we’ll have a set of long, horizontal “pincher” rollers to pry the gamepiece off the ground and pull it into the robot. We’ve found success with keeping these rollers at the same height, which will inherently allow intaking from both sides. While it is not a priority for our first iteration, we may eventually add extra vertical rollers to shuffle notes in from multiple sides.

Shooter

We’ve tried a few styles of shooter, including a slingshot (primarily as a way to speed up the note before feeding a normal flywheel shooter)

Since the notes are weird and squishy, there’s not a lot of resistance to compression. This limits your frictional force, and therefore limits your ring’s acceleration. If you’re going with a vertical shooter (like the kitbot) then we’ve found the most success using the grippiest wheels we can find, making sure they’re clean, and using multiple sets of wheels on different reductions/speeds to avoid slippage. If your note is slipping/ripping, you might need to slow down your first set of wheels.

If you’re using horizontal wheels/drums, then you can compress a bunch to get more friction, and then you don’t need to worry about slippage. We haven’t done as much testing with horizontal wheels, but it seems like there’s also more consistency from note to note.

Spin is good? Unless your shooter is built very nicely, and you shoot really fast, you’ll probably have issues with the note tumbling before it gets to the speaker. We didn’t have much luck with using grippier wheels on one side of our horizontal-drum-shooter to induce spin.

We did have success with holding a piece of box tube in front of the left side of our shooter so it smacks the note (making it spin) on the way out. It slowed the note down, but it stopped the tumbling. This was a demonstration that spin is good, but maybe not a viable means of inducing that spin. We plan on just spinning the wheels on each side at different speeds for our first robot.

For now, we’ve settled on horizontal drum flywheels on a pitch-adjustable shooter, mounted such that our intake can feed it at any angle. Here’s a clip of 3255’s shooter, which we are loosely basing our current design on.

Amp
We don’t trust amp shooting, we’re planning on having a small note feeder at the top of our robot. We’ll slowly “shoot” into it to feed it, and then shove it down into the amp. We’ll also score the trap and intake from the source with this mechanism

Climber
We’re going to start with an arm climb, similar to 3847’s prototype, but our primary goal right now is to get a robot up and running so our drive team and programmers can begin doing stuff with a physical robot. This means climber stuff is at the bottom of our priority list for CAD/machining/assembly, though we are including it in our layout sketches so we can add it later.

Here’s our primary sketch document as it stands:

Our team goes through weekly updates which helps us keep everyone on the same page about things, even on such a large team with many different projects and subgroups. It’s meant to be a combination of status updates and goals for the next week. If there’s any questions about anything in these slides, or this post, just let us know.

[ahartnet]

Wanted to share a couple of things real quick.

starting during the pandemic I’ve tried to send out a weekly vibe check during build season, and monthly or so in the offseason. I’ve been slacking the first couple of weeks of build season. I think this is especially useful for mentors to try to get a pulse of how the students on the team are feeling. I generally get a low response (15 out of 70+ on the roster), but I can at least be comfortable that I’m giving students an easy avenue to provide me input. Below is a summary of some of the stuff from this past vibe check

Also, here’s our 2nd weekly update slides: Pearadox Week 2 updates - Public - Google Slides

Hoping next post might have some more pictures/videos/CAD!

[ahartnet]

Some updates since last meeting:

Form taps

We started using form taps on the lathe last year. While you can do cutting taps on the lathe, we've been very happy with the form tap process. It does require a slightly different hole size, so we've added a sign behind the lathe to make sure that's known in case we run into a situation where we're gonna use a cutting tap somewhere.

Prototype - Pearadox 5414

Practice? We talking 'bout practice?

While we don't have enough of a robot together to do real practice (we're not hiding it, we're just behind the other teams) we are doing some drive practice. We have:

1. a base with thrifty swerve (needs to get fixed up this week)
2. a SDS base (intended to be what we put prototypes on…didn’t really happen this year)
3. last year’s comp bot
4. a Kitbot (which will hopefully convert to an Everybot in the next couple of weeks

This past week we did 3v1 and 3v0 matches (once we realized we were having some electrical issues with thrifty bot). You can see them in this smugmug gallery here:

PracticeMatches - Pearadox 5414

Some notes

• we had a mix of experienced and inexperienced drivers on the various robots
• I think 6wd robots without ground pick up will be at the source for awhile if they don’t get a way to practice. Even our experienced drivers had trouble lining up in the right spot between dividers. Swerve bot nudges were helpful
• when 3 robots end up at the source together it’s a bit of a mess
• we probably need to tweak our kitbot some, but if it’s representative of other teams initial builds, I think there’s going to be a lot of missed shots for clean up right at the subwoofer
• the amp and subwoofer are pretty close. I think there will be some collisions. And inexperienced drivers I think will be bumping into the subwoofer quite a lot.

Looking forward to a point where we have a robot and friends bring over their robots to practice with us too. For instance, our rookie friends 9478 (the Robo-Colts) came by a week or so ago and scored on our own field before us haha.

[ahartnet]

Are you feeling so far behind because of the great teasers and OA content?
Then don’t worry - this post is for you

We haven’t been posting too much, partially because we don’t have folks dedicated to it, and partially because we’re behind on some things. Seeing 5, 6, 7 note autos, and pictures and videos of robots coming together is simultaneously exciting, inspiring, and “oh my gosh we’re so far behind”-ing.

So - for any other teams out there feeling that way, don’t worry. Here’s some “not yets” for Pearadox:

• we do not have a base fully assembled yet (USPS in Houston really stinks and we didn’t get our modules until our Weds meeting after they’ve been in Houston since Jan 28th)
• we do not feel good about our launcher consistency yet
• we don’t have a design we’ll feel particularly great about taking to competition yet
• I’m gonna stop this list so I don’t stress myself out more

Now, that’s not to say we have nothing. We have been building up on a “alpha” bot that we anticipated learning some things from, but hopefully being able to practice with. We ended up dropping the top attachment with the rollers here to focus on building something up that can intake and shoot.

And we’ve got some of it built! But you’re only allowed to see the first 4 seconds of this video one

I definitely failed as a mentor in not scrutinizing some things a little bit more closely, and we’ve learned some important lessons (again) as a team:

• 2 runs of chain allows for a lot of backlash, especially in something you want to tightly control the position of
• we need adders on chain c-c calculations
• make sure you think about if your belt runs will get your desired rotation direction

Since we’re planning to rebuild some/all of this anyway, we think it’ll be OK. And I believe we can tack on a climber in the box if we need to take it to Belton. The point of building and running this was to learn from it - it’s just later in the season than we’d like. In the bag era, I would have been terrified.

But here’s some good news:

• I trust the students on the team, and know we have very capable students that can close the gaps we need to close this year
• team vibe, energy, and engagement has been great from my perspective as a mentor this year
• some of our newer students are starting to feel some dread recognizing that our super star seniors are graduating, and I think are trying to learn from them.
• we do have students working on CAD for our next revision with lessons we’ve learned from this

Anyway, we’ll hopefully have more to share soon!

[ahartnet]

Long time, no post.

We’re getting ready for district champs, and it’s been a fun season so far!

• Belton Winners
• Houston winners
• Houston Impact Award winners
• Top 2% World EPA
• Top 1% World Teleop EPA

Very proud of the team’s hard work and perseverance this year! We’ve been doing this with what was supposed to be our prototype robot There’s rumors of a new robot for district champs, we’ll see if it makes an appearance.

One area we definitely have room for improvement (like many teams) is scouting, both in having hype/excitement/engagement in scouting and in the accuracy of the data. I recently worked on some “accuracy checking” for our data. It sums up the notes we think were scored in our scouting data to compare to what is in TBA breakdowns. Right now we’re off by an average 5 notes per match from our Houston data. If you’d like to see our implementation of this, you can check it out here

We’re looking forward to district champs in a few days! There’s so much more we could share about what’s gone on between the last post and this post…but it’s been an exhausting few weeks.

Arguably one of our best matches (granted we had a pretty open field the whole time) - but we were 20/20 including some pretty long range shots: https://www.youtube.com/watch?v=Y59c1VyMiTY

A sneak peak of what we’re working on: https://photos.app.goo.gl/2VU9kH9NhXEGsKPE7

[chadr03]

Congrats on the season so far. Looking forward to seeing you guys in a few days.

Your sneak peek link is not working for me.

[ahartnet]

Always looking forward to competing with (and less so against) yall! You’ve got an excellent robot this year.

I think I’ve corrected the link now

[JamesCH95]

Your hard work and dedication is quite aPEARant.

[RicDV]

We MAY not be able to block Y’all…this time!!! LOL

[ahartnet]

DCMPs recap:

• We took our Beta bot
• we knew we didn’t need a ton of district points to qualify for worlds, so we were OK with taking a slight risk
• All day Thursday and in the morning of Friday we were facing various issues, primarily with brownouts
• Friday afternoon we finally fixed it and played 2 good matches, followed by our last match where we accidentally put in the wrong battery pre-match
• I think the robot worked reasonably well for playoffs, though we still have some work we’re doing this week
  
  

We had our “Pearascope” group using advantageKit logs (which are seriously incredible) and DSLog Viewer to determine where to focus our energy…on reducing energy use.

It was also cool that I could say “beta feels sluggish compared to alpha. Is that real or made up?” and they could come back with information like this from our logs:

And we could also use logs to see if our misses were robot or operator error. This is useful to know if we might need to adjust tuning for the field QR codes or something

I can not stress enough how helpful having these logs were to compare Alpha and Beta and see where we wanted to focus our efforts.

List of major changes we made at comp to address brownouts:

• Somehow even though our Alpha bot correctly used stator limits, we originally had supply current limits for the krakens on our drive
• we reduced our main battery cable length by like…a foot or more
• We discovered that our intake was drawing about 40A, due to a mix of a 1:1 gear ratio on a Vortex (more later) and wheels compressing enough that we believe it was making contact with carpet
• We changed our shooter motors from Vortex to our spare Krakens we brought for drivetrain
• at various points we tried essentially turning our shooter wheels off if we didn’t see the appropriate april tag and had really odd results of getting notes stuck in our robot - so we had to forego that entirely. I think I want to look into this more closely anyway at some point, though it might be an offseason thing.

The Vortex has a higher free speed, but lower torque. Our shooter design has a serpentine belt path, a bit of an upduction, and larger pulleys on our beta bot than alpha, all leading to a much larger current draw. So our alpha bot shooter idled at around 40 A total, and our Beta bot shooter idled at around 80A total. Switching to Kraken’s and implementing some stator current limits right after lunch on Friday as well as implementing the correct current limits on drive were probably the most important changes we made. I wish we would have caught them sooner.

FAQ - shooter belt path

We've gotten a couple of questions along the lines of:

We saw yall were using double sided timing belts on the shooter and were hoping to understand how yall get them to length (or are they preset) and how yall calculate the center to center distance using one big pulley vs multiple smaller ones

Another student can probably answer better, but the process I recommended to our students can be roughly described as:

1. create the belt path as a sketch, with lines tangent to all pulley diameters
2. sum up the length of all those lines and arcs
3. design in an idler tensioner that can be moved around
4. print the idler at larger/smaller sizes and/or print some of the pulleys 1 tooth more/less until we get a tension that we like

We purchase these belts from vbeltguys: D1000-5M-09 Double Sided Timing Belt – V-Belt Guys

That said, my understanding is that on Beta that process didn’t work well for us but magically the belt sizes we used for alpha worked well.

My understanding is that Solidworks also does have this belt path calculation capability. I think I’ve heard some friends import things into solidworks to get that belt length.

Onshape has been approaching a belt path length calculation, but isn’t quite there yet to my knowledge: https://youtu.be/ntjVvKzKDSo

[ahartnet]

There’s been a “trick shot” we’ve been wanting to do since mid-season. Motivated by wanting to one up High Tide’s video before champs, we took 15 min yesterday to film some attempts. Unfortunately it wasn’t 100%, but ~50% ain’t so bad. https://youtu.be/fXfGNT8BTKM

@jjsessa - we want to play a game of Crescendo HORSE sometime