Week 6 - Assembly (v2 Less Snow, More Problems)

It was going to be a challenge getting into Week 6 without posting
a properly spoiler image so I won't even try to bury the lede.

During Week 6 we were able to make up one of our missed days from Week 5. Did we catch up on all of our work? <Snape voice> Not hardly. With two weeks until our first competition we still have a few planned items to work on, though this week we've also discovered a few unplanned items. So grab your board and let's skate through Week 6!

Our calendar goals for Week 6 include testing our initial autonomous paths, getting drive practice, and iterating on the expectation that we'll start to see a few failures once we get the robot up and running. Generally speaking, one needs a robot to drive and test and tune, and we still don't quite have that robot thing yet.

We have made progress, including some catchup from our snow days last week. Next week we'll be fully caught up on shop time, though it will take a lot of focus and perhaps a time-bubble around our mills to produce parts at a rapid pace. (Any Mistborn or Sliders out there that can burn Bendalloy for us?) To help keep us on track and help us visualize our progress, we built a Gantt chart detailing the manufacturing, 3D printing, assembly, and order of operations for each subsystem on the robot. Every day we have been checking off boxes, but as more tasks complete we run into blockers or unexpected problems.

Williams F1 team used a similar looking spreadsheet to build their cars,
so we decided to make something similar to build our robots!

All that being said, by the end of Week 6 it really was starting to look like things were coming together. Literally.

Mentor Musing:

My Slack subtitle, tag, little quote thing under my name for the longest time was "Can I turn the wrench, please?". (It currently reads: "The laziest". Be good lazy engineers. Let your resources work hard so you don't have to.)

Every team has their unique coaches and mentors, and every coach and mentor operates just a little differently. My particular style varies depending on the week and our progress towards our weekly goals, but my core pillars are that students learn, feel ownership, and have some level of responsibility. In a Week 1 or Week 2 CAD session, I'll most likely be more loose - What do you think? How will that work? As we get to Week 4 and Week 5, the idea funnel will narrow and my comments will be more direct - Add a gusset here because we need to resolve this force. Move these standoffs so that we can fit a sensor. While direct, they also instruct on the 'why' of the imperative, which continues the learning process, and the final design is still the students - they are the owners.

When everything starts coming together, how much learning and ownership comes from spinning a wrench? Sure - there are 100% students who have never held a wrench or a bolt, and I want to make sure they get their hands dirty. But once you get into the 10-20-30 bolts all over different subsystems - I'll grab a wrench and help move things along. (Except for the unfortunate truth that we have literally one 3/8" wrench in the entire shop. It's fine, everyone just stand there and watch one person. Nobody walk over to the tool chest, there's most definitely nothing in there.)

Mentor CAD! By mentors who don't know how to CAD...
And stole 80% of this from the student CAD. At least we labeled the sketches...
But actually, we did this during our design pivot
because we really didn't know if it would work...

As the season progresses, the boundary of what I can and should pick up to help move things along blurs. I always want to go faster, but if I finish the last detail of a part, am I taking away a learning opportunity? But if that part finishes faster and unblocks multiple other designs or manufacturing or assembly, is that a worthwhile tradeoff? I think every year that I've mentored, I've hit some point during the season where I want to step in and take over <something> to speed up the process. Sometimes it's a gut check of - do I want to go faster? Or do we <need> to go faster? Other times it's teaching a new skill using the pattern: "I do, you watch. You do, I watch". And yet other times students will express their ownership and desire to finish the particular item on their own.

So, I have built a pneumatic system on a robot, I have taken a grind-wheel to a piece of superstructure, and this year I drew up a sketch in CAD. I suppose sometimes it is alright if I turn the wrench. Thanks!

Mechanical:

Parts drop! Last week when we finished the CAD, we emailed a sponsor to get several of the larger plates cut, and on Tuesday they arrived! We can start building for real!

Parts drop!

While building, we're learning that elevators can be complex, and while we checked motion and interference in CAD, we didn't quite check enough. Our modification to power the first stage chain run involved running a 1/2" shaft just behind the support crossbar. What we learned, very recently, is that the entirety of the second stage rigging and pulley setup needs to clear this axle. And it didn't. We immediately came up with a few ideas - smaller pulley (wasn't small enough, we would have to trim down the pulley mounting block), move the axle further away from the pulley (would require updating the chain attachment point and the lower axle, which would be challenging...), or split the shaft and figure out a capture mechanism. We're trialing the split shaft with a cantilever capture - it was the fastest and least destructive resolution - we can always make new pulley mounts and smaller pulleys in the future if the need arises.

Hex shaft and elevator pulley don't do well when
trying to occupy the same physical space...

That was really the major issue of the week. The rest of the time was spent manufacturing parts in a specific order (once everyone got on board with the Gantt Chart), assembling parts faster than they came off the mills, finding one or two more oddities in CAD, and losing some sprockets (more on that below...).

At the close of the week, the primary elevator and gearbox assemblies got mounted to the chassis. It looks decidedly more robot-like with parts on it.

Design:

All things are iterative, and though we finished the CAD, we started finding little spots where reality and CAD diverged. Some members of the Design team aided in assembly of their own or other subsystems, while others continued working in OnShape, updating bits and pieces as we found issues or inconsistencies. We still haven't run the robot, so we don't yet have specific thoughts on re-designs or improving designs for game piece manipulation or climbing.

Starting to lay out electronics.

Shortest design update all year!

Electrical:

Electrical team, in a bid to get a longer update than Design, started working on the competition robot, now that we had our planned and pre-cut chassis. In the team's current era (2022-present) this is the first time we had mounting holes and a manufactured panel for our electronics. But first things first - we had to remove all the bits off our wooden test board.

Electronic components mounted.
Everything perfectly balanced.

By the week's end, they had all the competition electronics mounted to the chassis, helped mechanical mount the elevator and gearboxes, and geared up for the planned 2 days of wiring. To really hammer home the point that the Electrical update is longer than the design update, our lead electrical student wants you to know that their favourite colour is purple.

Software:

Sadly, Software team lost the gam - er - robot this week. Fortunately, they still have our trusty 2024 robot, some //todo's in the code, and poking around with simulation tools like AdvantageKit. They popped into the shop a few times to check progress and make sure we didn't forget about sensors and cameras.

As the robot came together and Electrical stepped up to the plate, Software is now on deck, warming up, stretching out, and making sure we have all manual controls for each subsystem, but also each motor for initial bring-up, CAN id, and polarity checks. About two days into Week 7, we'll all be antsy to see this robot begin to move under its own power.

Week 6 Wisdom - Assembly (v2 electric bolt-aloo)

Ah, here we go again. Is assembly during Week 6 or 7 even helpful? From the incredibly small sample size of Open Alliance threads I have tagged, everyone's finished building and is just software tuning and drive practicing. (Ok, a few of them are breaking and rebuilding stuff...)

A soliloquy on keeping your workspace clean and your parts organized. At the very beginning of our elevator assembly, I told the story of 2023, where we lost some hardware from an elevator kit, and burned 4 hours of shop time searching for them and coming up with some method to replace them. This year we had three specific boxes with our elevator parts, and the boxes would be taken out only when needed, kept to a single workspace, and non-assembled hardware and bits would be returned to their boxes at the end of each day. We successfully assembled the mechanical structure, and successfully strung up the cascade section, then set about sizing the chain. We couldn't find the sprockets. We looked everywhere. Took most of the shop students over 2 hours before moving on, and ordering a new set. (We've been in the shop a few days since, poking around as we go through other bins, still no sign of them...) As a courtesy to yourselves, and to this story - please be mindful of your working environment, and your critical parts.

The aluminum tubes make it very easy to spot deformations,
but all-together, this thing does look good!

While performing the mechanical assemblies and whatnot, there are other standard procedures we attempt to guide and remind the team through. Last week I described different fasteners and anti-vibration hardware for the rigour of FRC competition. But those aren't necessary for the first assembly and installation. We generally begin with a dry-fit - put a few threads into each bolt hole, make sure everything lines up, make sure the pieces fit. Especially for purchased or tapped parts (but really, any part) we want to make sure the threading is smooth - we shouldn't be forcing bolts through machining or alignment issues. Along similar vein to forcing bolts, there is an ever constant reminder that tightening bolts is not a strength check. We sadly have a few tubes on the robot that have suffered the wrath of our well-fed brawny teenagers. Fortunately 3D printing is getting more accessible, so we can add crush blocks to limit the opportunities for crimped or warped tubes - but we can't always place them everywhere we want. Bolts should be firm and taut, not super tight.

Quote of the week:

Design Lead: "<other design student> is like, totally saving the day!"

Other Design Student: "I'm mentally dying..."