Week 7 - Application

It looked like we were really coming together by the end of the week

Week 7 was just a blur. So many times we try to compare to past build seasons - are we ahead, are we behind? While every season is different, around this time I'm sure everyone wishes they had more time, or were further along with the same amount of time. Our week consisted of more mechanical and electrical items crossed off the burndown list, more bugs found in the software, and by the end of the week we started running - and breaking the robot. We are not quite ahead or behind, we are right here.

Well, why not start with a quick celebration - our latest version of the intake works!! After some patience getting the software to do what we wanted, we successfully started scoring a few Notes in our mock Speaker at the shop. This was on our latest wooden prototype - though we were already getting set and planning for another local team to help us manufacture polycarb versions (Thanks 2046!). The software side of this was interesting - knowing when to be simple for testing, versus plotting the full final version you want to operator to be using. In testing, the preference is usually "just get it working!" - and we skipped that step. The software side of launching works with some assumptions and expectations - we assume the Note is contacting only the rear wheels, so we expect the front wheels to be clear, and expect the battery to have enough juice to spin them up to a given rotations per minute. Once we see that rpm, we feed the Note as fast as possible from the rear rollers into the front, and away it goes! Two-three steps that sound relatively simple took a few minutes to get working - under the pressure of the entire mechanical team - phones out and ready - looming and waiting. Personally, I was advocating for two buttons - one to spin the front, and one to spin the rear. This is my risk aversion and 'just do the thing!' mentality, but at the same time, I am impressed and growing in confidence of what the software team is able to accomplish.

Our first score with a "real" mechanism!

Alongside the intake, we really started getting the climber together, and really started running into some binding issues. For one of them, nothing a little grease couldn't fix - we'll have to keep an eye on it, but it appears to be reasonably springy at the moment. While we made progress, we didn't finish - our priorities require us to finish the intake and practice with software tuning the arm and wrist.

We expect not every team will be able to climb in week 1,
so it's important for us to be able to do so.

Our shoulder and wrist rotational joints use the built-in WPILib classes for TrapezoidalMotionProfiling. It honestly didn't take too much time to get up, and once we got our initial tune - I'm a big fan! As with everything else, we started slow, and will ramp up speed as we play with the system more and see if/where we have limitations. Once again I wish we made better use of Shuffleboard for tuning motion and setpoint values, but, that's all in the past now, and there's always next year. For now, the motion looks really nice, the commands are easy to use in both the teleop and autonomous modes, and we have time to continue improving the speed and smoothness. We have started building some autonomous paths, injecting the wrist and shoulder commands into the pre-defined path segments constructed in Pathplanner. Maybe next week we can try running an auto. Time will tell...

Later in the week we recieved the newly cut polycarb plates and re-assembled the intake - again. This also happened to coincide with our first scheduled outing to the practice field. I like having time at the practice field - for one, it's time available to us to drive and test and tune on a full sized field. For two, it's a forcing function - we gotta run what we brung. Work stops for the evening and we simply gut check where we are with the robot, what are the priorities we need to work on right at this moment, and my personal favourite - how can we break the robot? The shiny new intake was mounted on the robot, tested in the shop, then we headed swiftly over to the practice field.

Setting up on the practice field for the first time this season.
One week to go before competition.

I do like patterns and paradigms - one paradigm for Bearcats is dropping intake hardware on our first venture to the practice field. Once again, we didn't disappoint. Finally, here, on a real field we could practice picking up, manipulating, and scoring game pieces across a full cycle. We approached a Note, we lowered the wrist. We spun the wheels. And then.... ... Nothing happened. We uh, couldn't pick up the Note. Much whomp happened. This very strange situation occurred - in CAD, the geometry was perfect and everything fit and worked perfectly. In real life, the intake plate caught on the bumper and wasn't as low to the ground as it needed to be. Thanks to the full geometry of the shoulder and wrist, we could lift the shoulder a touch, which dropped the intake a touch, and viola! We were in business.

For now.

Throwing hardware around the field is somewhat amusing, but when it keeps happening and hex shafts start falling out of their bearings, it becomes a watch item. Then it becomes a 'theres definitely something wrong here' item. Apparently through all our iterations and testing, the intake width grew by 0.5". All our shafts were too short and pulling out of their retaining hardware. Fortunately, that is an easy fix for us.

Generally works better when the bolt isn't just... hanging out there...

As practice wore on, we noticed out intake motors getting warm. When I say warm, one student described it as 'You can probably fry an egg on the rear motor'. It was hot. Significantly hotter than it should be for the minimal usage we were putting it through. We noted it down as an operator issue - running the wheels when they were in the stow position and making contact with other parts of the robot. While we were happy cycling, one thing we hadn't yet done with out new intake was launch and score a Note (and check how accurate our home mock-up is compared to a real field). We lined up , Note in intake, and.... waited. Again. More software issues? The intake wasn't getting to the required rpm, so we never launched.

What gives now? We tried running the process again with no Note in the intake, and still the trigger didn't work. As we continued testing the wheels, once again a bolt withdrew, and for the first time of the evening, one of the shafts backed out of its bearing. Well let's get that back together. Try as I might, I couldn't pull the shaft and belt back into alignment with the bearing. A student came over 'Oh yeah, those belts are wicked tight, we had to use a wrench to force it into place.' Ah. Oh. Oooohhhhh. That explains a lot.

The sixth built-up intake getting fitted to the robot.

Hot-as-the-sun motor was straining against the impossibly taught belt. Cannot reach rpm due to severe mechanical drag in the system. Students (and mentor I found out later :-D) struggling in silence, and not broadcasting that they were encountering an issue during assembly. We found the root cause in our process - our latest intake design was rushed as we were dependent on another team to cut it for us, and parts of the review were skipped, namely the center-center distances on the belts. In a world of thousandths of an inch, 0.05 and 0.005 are very different numbers.

We loaded back up from the practice field, and the small group formed a quick huddle in the parking lot - We've learned some lessons here, but we're not quite where we want to be. We are okay though - we have some challenges, but we'll prioritize and tackle them one by one. Good words to tell a group of students, but it was Saturday night, 6 days from our first event, we needed another new intake and we couldn't manufacture it ourselves. To be continued...

No quote of the week, but fun hijinx in the shop.
Kayla and I had to hop on a late work call and locked ourselves in a closet.
A mentor gave us a helpful note on the door: "Work??? Retirement is so awesome!"