Opinion Poll: Proliferation of Prefbricated Parts

So Hopefully everyone tunes into the mild humor of my title; This isn’t meant to be a bashing topic, or a rant or any such thing, simply my personal opinion and concern that I have.

I’m a veteran of FIRST from East Islip Team 311 and West Islip Team 871 (NY), back in 2005, so while I haven’t been around since THE BEGINNING I’ve been around the block. I’ve been very lucky to be part of teams where the students are the real designers, idea pushers.

We (as mentors) have always attempted to guide our students to the best possible solutions, teaching them the various sciences (and arts!) of engineering. One of the hardest things we have to do is learn to let them make mistakes. Sometimes a mechanism looks like it will work, but then at the last minute wont. An experienced FIRSTer can probably stave off these kinds of mistakes early on, but should we? Don’t we know these mistakes because we ourselves have made them? Engineering is not “Do all the right things and make a cool widget” It’s about the hundred (or thousand) failed attempts that got you to where you were, that helped you UNDERSTAND why widget X simply can’t do thing Y.

So on to my real question:

How does everyone feel about the sudden Proliferation of Purchasable, Prefabricated Parts (P4 for all you DoD acronym lovers like me)?

Look at AndyMark and you can practically build a 100% functional compete-able robot just by buying prefabricated chassis, loaders, lifers, arms, grabbers. It caught my attention today, as I was looking for teflon track slides that my mechanical team has requested, that “loader assembly” and “Rhino Track” have been sold out; High demand for complex parts huh? So I looked further and discovered complete prefabricated assemblies for mech chassis, tank drive chassis, lazy Susans, swerve drive(?!).

Don’t get me wrong, Andymark is a wonderful supplier, and they have many useful things, I’m not bashing them. But is this really in the spirit of FIRST? Integration of COTS parts isn’t a bad thing, for sure, but when many of the “fun” achievable designs for HS students can be shelf bought, where’s the real fun? Sure, ball grabbers, and tank treads, lifty-things arent fancy for us professionals, they ARE intimidating design tasks, not to mention highly rewarding successes that our students can achieve!

I’m sure someone will point out that newbie teams (or simply less sponsor-gifted) teams will argue that “How else can we compete with big powerhouse team X, who has access to 5 7-axis CNC mills of professional company Y” , and I suppose I see their point. My team, West Islip 871, is one of those less-gifted teams. We don’t have the sponsorship of a massive machine shop, nor do we have a huge workshop of our own. Our robots are often ugly, and less functional than many others we see, so I understand the (honestly) feeling of despair some teams feel when they see a beautifully engineered and manufactured robot hit the field.

I argue however, that our ugly, maintenance nightmare of a robot, is in many ways more beautiful than some manufactured solidly-engineered masterpieces; Why? Because every part is student drawn, built and assembled. Does that mean we as mentors don’t do some strategic nudging? No. Of course we do. But with every robot we build, our students are so proud of their achievement. The glow you see in their eyes when they start explaining to the school board, or a fellow student, how this mechanism they helped design functions, is simply the most wonderful thing I have ever experienced.

How much do we take away from them when we start down the slippery slope of buying FRC specific, competition-made assemblies and bolting them together like erector sets? I for one don’t like where that path leads.

So again, this is my humble opinion, I don’t mean to bash anyone or anything. How does everyone else at CD feel?

I honestly think having so many pre-fabbed parts is nice. It doesn’t take as much just to have a basic robot be able to play the game. There’s still a huge ceiling though, as the best teams will continue to create game specific mechanisms that can insanely outperform any COTS mechanism. I don’t think very many teams would be able to get far without having a kit chassis, or without being able to buy COTS gearboxes.

There’s certainly an art to engineering, but I find that the students I work with don’t want to try and fail, if someone already knows it will fail. They’d rather try something that could potentially work. That may be just our differences in philosophies.

As an outsider, any of these robots will look cool and interesting. As a FIRST competitor and spectator though, I still find it sad that in the day of so many COTS options, teams still cannot field basic machines that can drive and move a game piece. I think some teams may need to reflect on their process, and determine whether buying a COTS intake solution and upgrading it would be more inspirational at competition than designing an intake that fails…

Before I begin, please everyone do not turn this discussion into a student-built vs engineer-build debate. I feel like this could turn into that VERY quickly.

That said, I have been a member of a team that did not have much support in money, knowledge, or engineers. We built very noncompetitive robots most years and were done quite frequently before lunch on Saturday. It’s not as fun as playing in the playoffs. These resources allow teams with less internal data to still be competitive. As a high schooler, I had NO idea how to get power from the motor to the wheel. I was absolutely perplexed. Now as a mentor I know, and these premade gearboxed help the new kids learn how it works. Kids that might not know the difference between a sprocket and a gear (they exist on almost every team).

I am all for it. Chances are the powerhouse teams are building it themselves anyway because they can do it for cheaper. And they have the manpower to do so.

Also, many of the so-called elite teams are also all student designed and built. You’d be quite surprised.

The prefabricated parts just makes the competition closer. It clumps up the middle of the pack to where one match can bump you up to 5th or drop you down to 30th. It really increases the drama of the competition and, in my opinion, makes things better. Even with the proliferation of off the shelf assemblies, they are rarely optimized and need some significant effort to make them more than just passable.

Because these prefabricated assemblies need to be optimized to fit your specific strategy it allows the effort to be put into making a great robot instead of just one that moves kind of like how you need it to. In my experience, the students have a better experience and have more to be proud of when they can build a great robot instead of one that simply minimally performs the task. We have used the KOP chassis almost every year it has been available, but it is rarely recognizable as such after modifications to make it suit our specific needs. It is wonderful not having to re engineer everything from scratch when there is a basic starting point to work from.

AndyMark.biz was selling gearboxes in “the old days” of 2005. FRC gave you a full drivetrain and instructions in the kit too! This discussion has been had many times. My team thought about the AM gearboxes in '05, and shelled out for them in '06… and I’ve been giving Andy & Mark my money every since.

I don’t think anyone wants to sell teams prefabbed game solutions (that wouldn’t be any fun!). What they do want is to make sure teams have a great FIRST experience… and for low resource teams that can be difficult. These teams need a lot of help. As FIRST has expanded, many teams no longer have access to as many engineers, machinists, hobbyists, and otherwise technically skilled people as they used to. The proliferation of FIRST suppliers like AM, VexPro, and design ideas like Ri3D & FRCdesigns has only improved overall robot goodness and the FIRST experience for students and mentors alike.

And if you want to do it the old school way, you still can! As a student, I really enjoyed the “Advanced Shop Class” aspect of FRC, and I enjoy passing that on too.

Before I begin, please everyone do not turn this discussion into a student-built vs engineer-build debate. I feel like this could turn into that VERY quickly.

Yeah I agree, I didn’t want that.

Also, many of the so-called elite teams are also all student designed and built. You’d be quite surprised.

Sure, No issues here. I wan’t trying to imply that beautiful robots WERENT student designed.

AndyMark.biz was selling gearboxes in “the old days” of 2005. FRC gave you a full drivetrain and instructions in the kit too! This discussion has been had many times. My team thought about the AM gearboxes in '05, and shelled out for them in '06… and I’ve been giving Andy & Mark my money every since.

I actually didn’t know Andymark went back that far. Gearboxes might be the straw that breaks the back of my argument.

Let me try to be a bit more specific, maybe I got a little too passionate and didn’t use the right words. I didn’t mean the mentors sit back and let the students screw up; I more meant for the mentors to teach concepts behind various designs and let the students apply it (and drop key suggestions when necessary).

The biggest reason (for my team particulary) is that for whatever reason, we can’t seem to build up and retain that organizational knowledge. We’re trying to train up new students as they move through the program, but there seems to be an upper limit to how much information we can cram into their brains before they cycle out of the program. It’s entirely possible (maybe probable) that we just aren’t doing a good job teaching. shrug

Our culture changing mission is served only indirectly by designing and building machines to compete in a game. Building teams (and, through those teams, new generations of informed, creative problem solvers) serves our mission much more directly. I want our students to learn that FRC robot design is less about designing components yourself than it is about learning what works and what doesn’t.

If there is an existing solution to your robot component problem, then your custom alternative should only be selected to go on your robot IF it is better. Having designed it yourself doesn’t not make it better, but improved function, reliability, cost, or readiness might do that.

I pretty much read OP’s post as:

When I was your age, I built robots with drill motors uphill in the snow both ways and had to remove the anti-backdrive pins too! AND WE WERE HAPPY ABOUT IT YOU WHIPPERSNAPPERS!

Seriously, COTS parts are awesome! Use them to your advantage. I don’t miss the old days.

As mentors we can help students avoid thousands of mistakes along the way, but you won’t prevent them all. There will always be failures, and the lessons we learn from them.

My personal philosophy is to correct every mistake I see when I see it, and talk to the students about the issue (and hopefully create a teachable moment). I want my team to fail at as high a level as possible. Your methods may differ as there is no correct way to mentor. I believe it’s impossible for FIRST students to not learn something regardless of how a team operates i.e. mentor built vs. student built.

As for your actual question, I believe prepackaged COTS solutions are a great thing for FIRST. It raises the floor and does nothing to limit the ceiling. Anything that accomplishes those two things is a great thing. I guarantee students on a struggling team will learn more from a functional prepackaged solution than they will from a non-functional “original contraption”.

Even if the prepackaged solution isn’t used, it may inspire ideas to make an original mechanism work. I can’t think of any downsides to prepackaged solutions.

There are a great many real-world engineering jobs that involve the spec’ing, selection, and integration of COTS components. Not all engineers work on the component design level.

I pretty much read OP’s post as:

When I was your age, I built robots with drill motors uphill in the snow both ways and had to remove the anti-backdrive pins too! AND WE WERE HAPPY ABOUT IT YOU WHIPPERSNAPPERS!
Seriously, COTS parts are awesome! Use them to your advantage. I don’t miss the old days.

There’s always someone. I did make an attempt to say that what you “interpreted” was actually NOT what I was saying.

Again, COTS Parts are great. I was more opinionated to the more frc-specific stuff, like prebuilt ball grabbers and such.

I don’t particulary appreciate the way you attempt to make me seem like a crusty old grumpy man… I’m actually not much older than my students. I also don’t want or expect them to design individual gears and sprockets and gearboxes and chains, etc etc. Cut me some slack here. Try to read what I’m actually saying. I don’t hate everything, I don’t hate COTS, I just see a pattern towards more complete purchaseable solutions.

I wholeheartedly apologize for using words that , I guess, made me sound like I expect students to “figure it out” and “deal with it”. That’s NOT where I’m going here.

My team is a low resource team. I’m sure the op knows us as we have competed with his team for many years. Until last year our team was hesitant to buy these cots parts either because we simply couldn’t afford them or because we thought they would take away from the experience. But after 3 years of fielding a robot that was only a drive train we decided to give some of these parts a try. They have changed our team allowing us to build working robots and finally not have to go to competitions fail a million times and come in last place. Our first working robot since 2012 was our offseason robot which took 2nd in a local offseason. The only cots speciality parts we used was the kit bot which we modified to use Mecanum wheels, competition robot parts roller kit, Rev gussets, and a banebots p80 gearbox.

We still had to manufacture a ton of parts and did stuff we never did before like tapping and getting stuff water jet cut. In fact the robot was 100% student designed and built.

Right now progress looks great. We have a articulating shooter that is 80% finished, a working drive train with pneumatic wheels. And CAD for a lifter. And we were just beginning week 3…

We are shooting a documentary right now on our teams journey obviously it’s not all thanks to cots parts but they have definitely played a role. Op we are going to premier the film sometime next fall at our school. You and your team should definitely come.

I do think there is a difference between a generic-use COTS robot part (a gearbox) and a game-specific COTS robot part (an intake). That may be the distinction that upsets some people.

What’s wrong with crusty old grumpy men!!! I’d be out a job without them…

I did say that was how I read your post, not how you meant it. Funny thing about meaning (and much like my sarcasm), it’s not always conveyed in the way we wish it to be.

I’m all for more pre-made assemblies and more robots in 3 days! Bring it on!

Different strokes for different folks. It’s all about the evolution of the individual team and how far mentors can/want to take them.

Who benefited the most. The team who bought a swerve drive from the vendor? Or the team that engineered or re-engineered the idea?

Yes, cots equal the playing field. I’m fine with that. But it’s the journey not the destination. Don’t buy all cots at the expense of learning.

The interesting thing about the intake mechanism AndyMark is selling is that while it seems like a game specific thing, if you look at the last ten or so games for FRC, you could use it as is in probably 7 of them as is, and the other 3 (2007, 2011, 2015) you could modify it a bit for use in, say, a roller claw. So while it may look like something game specific, really a roller intake is just another common component of a robot. That’s a big difference in selling that versus, say, a variety of 7, 8, and 10" flywheel ball shooters.

Nicely said. Maybe that’s what I was getting at all along?

The only cots speciality parts we used was the kit bot which we modified to use Mecanum wheels, competition robot parts roller kit, Rev gussets, and a banebots p80 gearbox.

Yeah, Those kinds of parts I’m all about. We buy COTS parts too, like gearboxes and the like, linear slides etc etc. I think part of the problem is that I’m still new to the whole “leading the organization thing”. I’m just a software guy (Blame it on the software right?), so I’ve got relatively little experience designing physical stuff. But now I’ve slipped into a leadership role for the team itself and I’ve got some freedom to actually make choices that can affect where we’re going.

I won’t lie, that still kind-of scares me. I’m starting to ramble – and I know it – Sorry about that. I suppose I wanted to see how other teams approach the problem. I like to hear from people who have lots more experience than I do and try to understand how they think.

Thanks for the discussion so far everyone.

The product of FIRST is not necessarily the robots. It’s the teams behind the robots. FIRST wanted to expand their program almost two decades ago beyond its very narrow scope of industrial partners adopting schools and turned the relationship almost entirely upside down.

There is some sort of spectrum that every aspect of FIRST operates on. While expansion definitely has its own spectrum and I have my own opinions on it, let’s talk about a side effect of the current level of expansion.

You may believe that the line has been crossed in terms of the relationship suppliers have with creating specialized products that cater to FRC teams, and you are right to have an opinion on it. One of the great things about FIRST is that I really think it can be all things to all people. The goals one team has may not just be different in sheer scope or difficulty, but may diverge at an even earlier fork in the road. Some teams do not operate under the idea that FIRST trains the next generation of engineers. Some teams recognize it as an opportunity to show kids the potential of STEM fields. Some teams find FRC to be one of the most high profile and effective programs at the high school level that can build teamwork and leadership skills. Others find it a great enabler of community service.

Sure, FIRST has a crafted mission and vision for its program, but teams should also have their own mission and vision for their own program. The meaning of participation in FIRST is whatever the participant defines it to be, which is why it can be all things for all people.

In terms of the spectrum of the relationship suppliers have with FRC teams, they really are not crossing a line for me until they are boxing up MCC kits and selling them as a separate SKU. Even then, the construction of an FRC robot can very well be a tiny fraction of the execution of the entire technical division of the team.

I’ll say this. I don’t think many teams buy COTS without a good reason. I know that on 68, we finally made the switch to COTS drive transmissions in 2014 after making them custom every year prior. Could we have made custom transmissions again that year? Sure. So why didn’t we? Because that was something we knew we could buy a reliable COTS version of, and doing so would allow us to focus our efforts on other custom mechanisms and get parts out of the CAD lab and into the shop faster. That year, in no small part due to the amount of time we saved and the manpower buying COTS freed up for other things, we finally fulfilled our perpetual goal of having almost-identical practice and competition robots. Did it prevent some students on our team from learning about transmission design? I suppose, but it also allowed them to be more involved with learning other things and it improved our build season schedule and performance overall.

I think that was a trade worth making, and I suspect other teams that seem like they should be able to get by without as many COTS parts are evaluating their choices similarly. They aren’t shorting their students out of the chance at learning, but rather allowing them to move on from systems that would otherwise consume a large part of their build season and inhibit having a functional or competitive robot. Whatever stage of development or competitiveness a team is at, they can move one rung further up the ladder and help their students learn something new by pushing their competitive ceiling with COTS parts.

I think this is a common feeling on the matter.

I think this is a fantastic discussion to bring up. I do think it is only a matter of time before companies start selling kits to make game specific systems (intake, climber, shooter…) that could be combined to build a complete robot. The past few years have brought companies building complete robots, providing CAD drawings and selling kits of the more challenging to produce parts for these robot systems. The next step is for companies to explicitly sell robot systems along with step by step instructions.

I think it would be fun for the company if they could sell them for a significant profit. :slight_smile: