Welcome to our thread!
2025 Links
4192 is excited to join the openalliance for the 2025 Season! We’ll be competing week 2 at Plano and week 4 at Fort Worth.
Crescendo Recap
Highlights
- Received Impact Award at Plano
- Ranked 4th at Amarillo
- Received Team Sustainability award at DCMP
We finished the season ranked 39th in FiT District (3 DCMP playoff wins away from Champs).
I’d also like to congratulate our own FTC team 12791 for winning the FTC Inspire Award and representing team USA at FIRST Global!
3 Likes
NTX Recap
At NTX, we ran our 2024 robot unmodified from the 2024 season. We had several issues and mostly played defense as a result.
Thanks to 3676 Redshift Robotics and 6369 Mercenary Robotics for inviting us to your alliance!
Problems & Solutions!
- Left shooter motor wasn’t working (pre-event)
- Reconfigured Spark Flex CAN ID, still don’t know why it reset.
- Intake rollers didn’t have enough friction to intake note (pre-event)
- Moved redirect plate
- Solution: switched to using newer notes
- Shooter flywheel bearings fell off (Qual 5)
- Solution: don’t shoot without shooter angled up
- Elevator carriage rigging lost tension (Quals 10+)
- Retensioned using CAM (worked temporarily)
- Solution: we gave up on using elevator to focus on fixing our intake
- Intake pivot chain lost tension (Quals 10+)
- Tensioned using original slots (Didn’t provide enough tension; quickly failed)
- Removed chain link (chain was too small)
- Ran zip-tie to tension chain (zip-tie snapped)
- Solution: Only ran intake at start of match, then switched to defense
- Intake pivot mounts bent after truss collision (Elim 2)
- Solution: Hammered Rebent intake into frame perimeter
- Batteries constantly browned
- Swapped batteries
- Verified condition with beak before match
- ?
- ???
Lessons
- Lower bumpers = better defense
- ALWAYS include chain tensioner in design
- Or, don’t use chain
- You can overdo battery mounting (our batteries were almost impossible to remove; 2 4” strips of velcro + a strap above made the swaps a pain)
- Simpler = better
- Almost every mechanism broke, costing us tons of time between matches.
- Our elevator was never used, so having a better intake + shooter would’ve probably improved our performance
- Basically just Karthik’s second golden rule
We had a blast at NTX, thanks to all the volunteers and teams who made the event possible!
3 Likes
Offseason UTB Intake
We disassembled our 2024 intake after NTX. We kept our shooter/elevator together for reference/part storage for 2025, but our intake is back in the cabinets.
Rapid Prototyping
To practice rapid prototyping, we built an UTB intake, and it mostly worked!
We used both 1/5” plywood and 1/8” hardwood with success.
Prototyping went well, but we made a few mistakes.
- We prototyped with our robot on the cart, which made the note path unrealistic. On the ground, the intake would burn up the note (and carpet).
- We originally didn’t cover our rio, and potentially got more note powder in the holes.
- The chassis tubes on CAD were different from real life, so our mounting didn’t work that well.
- When we cut the first pieces, the laser didn’t cut through the entire part in 1 pass.
Final Design
how do you like to see prototypes on Chief?
- GIFs
- Youtube
- Google Photos
- Other
0
voters
1 Like
2024 Robot Damage
We disassembled our drivetrain and inspected the damage from 2024.
The damage wasn’t too bad this year: our drivetrain tube did bend at the bumper mounts, but the damage was contained to a small area.
Overall, while we’re happy with our current levels of damage, we’ll make a few improvements for reefscape:
- Switch bumper mounts from rivets to bolts
- Run more crush blocks between bumpers and chassis
- Secure bellypan with bolts instead of zip-ties
- Add more bumper mounts
4 Likes
Underrated meme
2 Likes
Pic: Final 2025 Drivetrain
Based on the leaked reefscape manual, we designed our 2025 drivetrain. For 2025 we’re planning a 28”x28” chassis.
The bumpers are composite parts, so hiding them when looking “behind the bumper” should be easy.
Why?
Thanks to onshape, drivetrains are easy to cad, so having a configurable drivetrain ready before the season doesn’t cost much time. More importantly, the drivetrain can be a reference for good practices and file organization.
2 Likes
Spreadsheets/Documents
I’ve made a lot of spreadsheets/docs for FRC over the years, here are some that you may find useful!
Grant Database
Grant Database
This sheet, developed by teams 353 and 358, has helped us a ton with submitting grants/sponsorships (and my college apps lol). This is the same as the original, but with more formulas (google sheets is the only programming language I know).
Because the shared sheet is not actively updated, the dates/requirements may be off. I also removed some that were not relevant to us, but adding new grants/sponsorships is not too difficult. Make sure to check with the original and update any incorrect ones (:
Thanks 353 and 358 for the awesome resource!
driveCalc
driveCalc
This is just data from reca.lc in a spreadsheet comparing different COTS Swerve ratios. The scores are arbitrary and don’t work. The spreadsheet assumes a 118 lb robot (our 2024 weight) with 15 lb of auxiliary weight. I’ve used this sheet to compare ratio choices between teams and motors (L4+ Vortex on a 40lb robot is the future).
REBOUND History
REBOUND History
This is a document showing the progression of our DunkTank robot REBOUND’s version history. It creates some pretty cool graphs, and highlights “trends” (procrastination) in our design process. I’m definitely making one for our 2025 season bot; we’re at version 30 or so already (:
Amarillo Scouting
Amarillo 4192 Flower Mound High School Jaguar Robotics Week 5 pre-scouting
Here’s our prescouting before Amarillo. We used QRScout for data collection, but quickly realized quantitative data was mostly useless (vibes>data). The cool thing about this is the “PlanoData” Sheet, where I calculated that, through match 27, the alliance who collected the most midline notes won 89% of the time. Interestingly, we were on the only alliance in this dataset to win a match with only 1 midline note (statbotics gave us an 8% chance of victory before the match). We had a wild numerical season: statbotics went 4-8 at our first event, 10-2 at our second (with a streak of 10 matches) and 12-0 at dcmp (22 correct qualification match streak).
Post Amarillo Scouting
For our first pick at Amarillo, we basically just used vibes (the data on teams was too close to determine the “correct” pick). Afterwards, I wanted to see how we could avoid making the same mistake in the future. Here are some of my findings:
- We were cooked in our position, 4153 was a good pick and did well in playoffs
- Autonomous route scouting would be useful
- removing each team’s worst match would have given us better data (rampage was phenomenal even with 3 dead matches; honestly my favorite texas team)
- My vibes > @Varunaboys2 (but he predicted alliance 1-6 1st picks/captains at dcmp night before, so maybe Amarillo is cursed)
Ultimately, we switched to Lovat for dcmp, which helped us improve for playoffs/alliance selection.
FiT Event Planning
FiT Events Planning
I created this spreadsheet to find the “easiest” event for us in 2025. It’s not particularly useful, but it does provide some interesting insights. Rank (opr) and Trev Rank are both for 4192 (where our opr ranks, and where I rank us. If I remade this today, the trevrank would be <7 for all events)
4192 Open Alliance Google Doc
This is how we cook up our open alliance posts. Google’s headers automatically copy into markdown, so no extra work is needed. There is also a (not well hidden) easter egg in the document (:
I hope some of these turn out to be useful! I love making spreadsheets, these are some of my favorites.
4 Likes
I would love to see more about the graph/how it works.
Looks awesome, love the excel work (and may need to steal some inspiration for upgrading the grant database.)
1 Like
The process is pretty smooth.
I just take the largest version number from a day and copy it into the sheet.
There are a lot more days in the season that I would’ve thought (especially when you start CAD 2 months early), so later on I’ll update it with a weekly option as well (the chart can become messy).
To save the graph, you can use sheets’ “download image” option, which lets you save as a png or svg.
Happy kickoff! Over the coming days, we’ll post our strategy, would’ves, should’ves, could’ves list, and a few other things.
Good luck in REEFSCAPE!
2 Likes
i’ve got a lot of thoughts.
2% of build season gone, 98% left!
Today we met at our library to discuss our team goals, watch the reveal, read the manual, and consider effective strategies.
We created a needs/don’ts list, but we misread the manual and didn’t realize that the alga goes to the opposing human player, not your own. So, we’ll reconsider our list of objectives.
| rank | Needs | Don’ts | tbd |
|---|---|---|---|
| 1 | auto leave | opposing side auto leave | preload coral score |
| 2 | alga only | coral | |
| 3 | alga ground intake | deep climb | |
| 4 | alga auto intake | ||
| 5 | park (june) | ||
| 6 | shallow climb |
Our analysis of this game found a couple of things.
- Scoring alga & coral is too complex
- RPs are easier than in previous years, with 5 being available with just a drivetrain. (luck in match schedules could have huge impacts)
- Depp climb is difficult (maybe 2019-style arm climb?)
- Shallow climb still requires another shallow climb + park for rp.
- Potential L2/high ratio against defense.
- This game is difficult, but a simple robot can do well.
With an alga-centric robot seeming unlikely for us this year, we’ll start planning which field objects we’re constructing for 2025. The reef is necessary, but the coral dropper and barge are questionable for us (we’d rather prototype climb on an actual field). As of now, the processor seems unlikely.
I’ve done some krayoncad, expect that & some sketches later tonight. I wanted (anti-3DM) to do a 1 dof robot before game reveal, but that seems unlikely given the performance boosts of a second dof.
Kickoff Concepts
“Just watch ri3d bro”
Double-jointed arm
2dof (without ground intake).
Pros
- Simple
- Scores levels 1-3
- No reorientation required after ground intake
- Ground intake addition could be made between Plano (week 2) and FiT DCMP (week 6), allowing us to see other intake designs.
Cons
- Little arm experience
- Arm has two joints
- transfer could be difficult to get consistent
- May require second limelight (one for scoring alignment, one for intaking alignment)
Coralvator
“He made me experience high tide!”
This design uses an elevator & wrist to score in all four levels. Similar to 3847/4414 in 2023, it could be another option.
Pros
- Greater electronics access than the double-jointed arm concept.
- Experience with elevators (past two seasons).
- Potentially more-reliable transfer
Cons
- More complex
- Heavier
- Harder to add ground intake
- Harder to add dual-sided source intake
4 Likes
I don’t know why but the correlevator really looks like it should be a double pendulum to me
Democracy!
What autonomous are you planning to run in 2025?
- Mobility
- Score preload coral
- Score additional coral from human
- Score algae in amplifier or stage
0
voters
Where are you planning to score in 2025?
- Algae - Net
- Algae - Processor
- Coral - Level 1
- Coral - Level 2
- Coral - Level 3
- Coral - Level 4
0
voters
What climb are you planning to run in 2025?
- Park
- Shallow
- Deep
0
voters
Is it a Trap?
(Alt: my Reefscape hot takes)
Is Level 3 Coral a TRAP
- Yes
- No
0
voters
Level 3 coral requires algae removal (2 algae per coral). In order to score level 3 coral, we’d need to consistently be able to both
- quickly remove algae from reef
- keep algae out of robot
These two tasks could require a lot of prototyping, which may be different from the final field.
Is Shallow Climb a TRAP?
- Yes
- No
0
voters
Shallow climb is worth 6 points, requiring another shallow climb & a park to reach the barge rp. 6 points is also most likely inconsequential for playoffs (a park is more efficient), so it may be worthwhile to go deep or go home.
Is 2-Cage Deep Climb a TRAP?
- Yes
- No
0
voters
Dual-caged deep climb may also be a trap. While climbing on two cages does seem easier, cage alignment may be difficult to consistently climb (without swinging). Prototyping and testing this system would also be a challenge.
Is Climbing a TRAP?
- Yes
- No
0
voters
Is prototyping & testing a climb really worth it? If deep-climbing is common at our two district events, it may be worthwhile to just park instead of designing a deep-climber. Potential designs right now (arm latch + drivetrain rotation, dual telescope climber for two cages) may be too slow to comfortably run in playoffs, and the traffic near the barge during playoff match end could take more time.
I’m curious to hear other opinions on potential Reefscape traps.
1 Like
Archetype Development
Algae Removal
We added a fixed roller claw to the double-jointed arm concept, and the geometry seems to fit. This robot is able to accomplish most reef tasks (except for L1 and L4 scoring). The ball can only remove the algae between levels 2 and 3, and seems to be capable of scoring into the processor (with assistance from coral intake).
If we chose this archetype, we won’t pursue level 4, as we’d need either
- a third arm joint, or
- a higher cg,
neither of which are a good idea for this game.
1 DoF Elevator Concept
This elevator only uses 1 DoF to score in l2, l3, and l4 on the reef with human station intake. However, compared to the 2DoF arm, the elevator is still more complicated.
2 Likes
Other Updates from Day 2
We won’t meet in person until Tuesday (expect a mega post that evening or Wednesday), so for now we are planning out our next few weeks.
Field Elements
Here are the field elements we’re looking to construct:
- 1 reef side (more as needed for auto; we’ll look over different designs tomorrow)
- Coral station (we may redesign to use more accessible materials)
- processor? (dependent upon goals; probably reuse amp)
Barge doesn’t seem too useful yet; we’ll probably wait to see what other teams come up with for climbing.
Prototypes
We’ll start designing prototypes soon, here are some things we’re wanting to test out:
Coral Intake
(Cranberry Alarm end-effector prototype)
We’ll be looking to intake coral from the station and potentially ground, so there’s a few things we need to test out. The cranberry alarm end effector seems to work with our design, and may be able to remove algae from the reef.
- How much compliance should our intake have?
- How many sets of rollers do we need?
- Can coral be vectored into an orientation/position?
Algae Intake
(our double-jointed arm design with a fixed algae reef intake.)
We’ll mostly be moving algae from the reef, so the testing is focused on finding ways to package an algae intake.
- How many sets of rollers do we need?
- Do we need a compliant or non-compliant intake?
- How can we hold algae through a collision?
Climb
(our 2024 climb used passive hooks with surgical tubing to return to position)
To determine whether a shallow or deep climb is more valuable, we’ll need to test different latching methods.
- Can we use a telescoping tube to grab the bottom hole of the cage?
- Can we use a passive latch to grab the cage tubes?
- How strong does the passive latch need to be?
4% of build season gone, 96% left to remove!
Design Update
Last night, we developed a mastersketch for the coralarm concept. It should be capable of scoring into L2 and L3, while also removing algae from the reef.
The arm uses the same position for L3 and L2 scoring, allowing us to rapidly switch.
This design seems to be pretty competitive. It also saves room for a potential ground intake later in the season, if ground intakes are needed.
Here are the main problems with this robot:
- Difficult to add L4 scoring
- Small intake (~10” wide)
- Potential CG issues?
- Arm position issues?
- Arm speed issues?
3 Likes