Did your team overreach with your 2019 robot design?

This build season, did your team overreach or underreach with your robot’s design and strategic goals?

  • Overreached
  • Reached just the right amount
  • Underreached

0 voters

Bonus discussion questions:

  • What do you wish you had done differently?
  • Is it just me, or does it seem like the majority of teams overreach?
1 Like

NOTE: This is my opinion and my opinion alone. This may or may not be the view of my team.

I’d say we mainly overreached in our intake. Our climber was also known as our drivetrain via Climb By Yeet, our manipulator was really well made and designed, but our intake for balls was one of the sources of many problems. Our proto-hatch-floor-pickup, which never made our robot, was nixed because of various electrical interferences. (Remind me again why we can’t block off all access to the battery lol) We never thought on integration for our intakes until it was too late and 2/3 of the robot had been built.
I, for one, wish we had gone with one of the original original hatch and ball intake prototypes, which had the problem of integration well thought out. (Yes, I did lead this prototype. Yes, I’m biased. And I’m trying to keep the bias from interfering) The main problem with it, however, was its size and shape, as well as that annoying elevator, which is why it was eventually nixed.
I’m not saying that we would’ve overreached had we built that, but I’d say that we wouldn’t have overreached quite as much and invested so much time into the proto-hatch-floor-intake design.

1 Like

We tried passthrough.

We didnt have passthrough post week 1.

We could of climbed if we didnt do passthrough (elevator width/weight), which was arguably more viable than our passthrough.

the elevator was sick and it worked 100% of the time at least, to its credit. Best one we’ve done yet. Came a long way from our 80/20 elevator in 2015.

3 Likes

Honestly something that I wish we would have done differently was use neos since the beginning of the season, we had cims up until IRI and I think the introduction of neos really helped us at IRI this year but other wise we stuck to our plan of keep the robot simple and easy to repair

2 Likes

In my team, we made a very well level 1 hatch bot, but there were some things I wish we added/changed to it.

  1. Climber

A big part of deep space was climbing. We originally did not design a climber, because we did not see it as a top priority. But, when we got to our first competition, we realized that climbing can play a huge role on your rankings. If we had designed our climber before our week one event, we probably would have ranked higher, and forced our second quarterfinal match to a third quarterfinal match. Because of that, our team decided that we should make a climber before our week three event, so that we can rack up more district points. Unfortunately, we were not able to. So, we went to our week three event with the same exact robot as we had in week one. Our week three event was 100% focused on climbing for rankings. Because we did not have a climber, it kind of hurt us on where we ranked. We ended up getting picked first round by the seventh alliance, and made it to the finals. We lost in the finals in 2 matches.

  1. Ability to go off of level 2

When we got to our first competition, we had this debate about if we should take the risk of going off of level 2 or not. We decided to not go off of level 2, just to be safe. But, the fact that we did not go off of level 2 kinda hurt us ranking wise. I believe that, we could be ranked much higher, if we went off of level 2. In our week three competition, we again decided that we will not go off of level 2 just to be safe. I firmly believe that we could have been ranked much higher, if we went off of level 2, just because most of the matches were low-scoring, and the scores for both alliances for each match were pretty close to each other. IN DCMP, we finally made the decision to go off of level 2 for one of our practice matches. We did, and we did it safely, without breaking anything. If we decided to go off of level 2 previously, I believe that, we would have got more district points.

  1. Switch from mecanum wheels to tank drive wheels right before nedcmp

We had a mecanum drivetrain for our week one and three competitions, so that we could have easy placement and intake of hatches. We also were not a huge target for defense. But when we got to DCMP, teams were asking us to play defense during quals. Since a robot with a mecanum drivetrain can be pushed very easily, we were not the best at defense. Because of that, we were not very attractive for alliance selections. We had an amazing drive team, and I believe that, if we used tank drive wheels for nedcmp, we would looked attractive for alliance selections, and possibly be picked.

Our match mechanism was criminally underdeveloped. We tried suction right off the bat, seeing as we didn’t use pneumatics, and we never really dove too much into testing it properly. After a poor performance at our first district, in reguards to the mechanism, changes to it were made. Despite these changes, it still underpreformed in comparison to the rest of the robot. I 100% wish we had the compliant/roller intake.

It’s not just you.

Where do I begin? Our intake was far too heavy for our elevator, which was geared to some ridiculous unattainable speed. The wrist joints involved super tight chain runs that were impossible to get your hands into and doubly impossible to get proper tension on. No one planned out electronics, resulting in them being a mess. We never properly thought through our HAB 3 climber, so it never went on and probably wouldn’t have ever worked. All in all, planning was just sloppily done.

This is completely my view and probably not the view of the team

Although we did go for the turreted elevator, I still believe that we under reached because many parts of the robot were slacking. The first thing that I think we under reached with was the elevator. Instead of doing a custom gearbox for the two 775s, we used a versaDM gearbox going into a versa planetary. This was primarily because it would have taken more time to design and make a custom gearbox. This led to many issues, the elevator was very hard to back drive which made resetting the robot very difficult, the elevator was also slow and inefficient. We also didn’t take the time to solve why the elevator was very hard to move which we eventually found out was from the pulleys that held the rope rubbing on brackets meant to keep the rope captive.

The next area in which we under reached was the turret. Instead of taking the time to properly make a large gear which would go on the turret, we took the easy way out and stretched a belt around the outside of the bearing and used a 775 on a versa planetary with a delrin gear going to the bearing with the belt. This lead to issues like the turret would occasionally slip teeth with high acceleration and deceleration.

The third system in which we under reached was the climber. We went with a four bar mechanism actuated with two, 250 pound gas shocks as opposed to a more complicated PTO from the drive train. This led to a few problems like the climb being unreliable with one side sometimes going up faster than the other which led to a failed climb. Even when the climbing mechanism did work, it was far too slow to be competitive with.

Feel free to ask me questions or comment on what I said.

1 Like

In 2018 our robot had a swerve drive, could score cubes everywhere and had a buddy climb system that could lift a partner robot. It took us a long time to get all of that working to the level we wanted and we barely got into Detroit Champs.

In 2019 our robot had a 6WD, emphasized our level 3 climb as our top priority and scoring low everywhere else as our next priority. We ruled out scoring on levels 2 or 3 of the rocket on day 2 of build season. We finished 3rd in the Indiana district and captained an alliance at all of our district events this season.

Simple works folks. You can always make it more complicated if you need to :slight_smile: Just get it working first. Evaluate the easiest way to maximize RPs and win matches and do that early / often. It worked out pretty well for us this year.

10 Likes

Coming from a team that, prior to this year, had never once made finals at an event, had never once made world’s, and only once made DCMP in our 12 year history, I implore other teams to not undercut their own potential, just because you have a history of doing so.

All too offten I see teams with ample ability to do better, yet fail to take advantage of their opportunities, just because they fail to set their goals higher year after year. 2412 used to be a prime example of why you shouldn’t set low goals: if you never strive for anything better, you’re never going to achieve anything better.

In one year we had the single largest jump in performance our team has ever seen, and I attribute that almost entirely to our high goals (making world’s and making it to the finals), and our background commitment to always seek out the best solution to a problem.

This does not mean blindly set unrealistic goals that will only make yourself overextend! But continuing to have crappy work ethic or methods year after year just because “that’s the way we’ve always done it” is a recipe for a failed team. You will fail to win. You will fail to inspire the community arround you. And you will fail your greatest responsibility: to educate your members.

For 3 years I watched students join and leave the team learning little in comparison to what our students learn today. That’s why I kick myself every day. Not striving for better trainings, or setting up better practice, let students go through FRC and learn magnitudes less then if we had strived just a little bit higher.

Please. Please. Set goals. Even if they are as simple as “make playoffs” or “host a CAD training.” As long as you are always making steps forward to be a better team, and a better person, you are successful. Giving yourself a target for improvement gives your team insentive to improve. And then next year you can aim higher. And again.

I’m reminded of one of my favorite quotes of all time, which maybe some of you guys will appreciate:

"The greater danger for most of us lies not in setting our aim too high and falling short; but in setting our aim too low, and achieving our mark" (Michelangelo)

10 Likes

Coming from a very old team that is pulling itself out of a deep rut, I cannot agree more.

This is invaluable. Set goals.

In 2018, our most successful year since 2010, we set some relatively high goals based on simple, but high, targets. We had a deadline, we had a goal, and we did it. We got to champs, not through the waitlist, but through our perseverance and work.
As John F. Kennedy said,

“We choose to go to the moon in this decade and do the other things, not because they are easy, but because they are hard, because that goal will serve to organize and measure the best of our energies and skills, because that challenge is one that we are willing to accept, one we are unwilling to postpone, and one which we intend to win.”

Well, we build robots, not because it’s easy, BUT BECAUSE IT IS HARD. (And fun, but that’s beside the point)
We set ourselves a deadline, a goal, and a plan. Our goal was simple: Build a working robot that can play this game decently well. Our deadline was, of course, Stop Build Day. Our plan was uh… something along the lines of “Make it up as you go.” We succeeded in our most successful year since 2010, going to championships as a result of skill, luck, and hard work. (Wildcard as third robot on the finalist alliance AKA dumb luck)

However, I am not saying to go above your team’s capabilities.

Know your limits, and don’t try to go past them. I’m reminded of an article in JVN’s blog:


Look at the team in this story, but more than that, make sure you push your limits, but don’t exceed them.
In 2016, possibly one of our worst years as a team, we decided to try for everything, resulting in building a shooter that had some ridiculous idea that the ball was supposed to go everywhere but where it aimed (We barely ever used it), trying to climb (That got scrapped before it made the robot), and some sort of super-duper-get-across-the-defenses arm/intake. The arm worked, but barely. Our drivetrain was the most successful part of the robot by far, and it was the only thing we tried to build within our limits. Other than the drivetrain, we tried things we had no experience with, and we tried to build past our limits. The result? Rank 35/60 in one regional and rank 33/65 in the other.

3 Likes

To understand the Pandamaniacs this year requires studying the Pandamaniacs last year. Gross overreach in design (the goal was a double pneumatic lifter) compared to our ability to execute led us to wind up bagging a robot that could only run the exchange. We designed a lifter with our withholding allowance that did get the robot off the ground…once…on the practice field. Combine that with dissatisfaction on the student/mentor balance, and we were due for a reboot of our process.

We really took the MCC concept to heart this year, inspired by Spectrum’s MCC guide, and opted into the Kickoff kit drivetrain for the first time in several years. My day-one impression of an MCC wound up looking similar to 118’s Everybot, and we were giving a lot of study to a wide-chassis version of it when Spectrum dropped their MCC vision that used a single-joint arm instead. Single-joints are not something I’ve had massive success with historically, but their approach was a lot sturdier than past attempts and enabled more reach and a path to the second level of the rocket. So we shifted our design focus towards that.

One thing the Spectrum MCC didn’t really flesh out was a hatch panel system; we poached ours from 95’s Onshape and made a version from our favorite locally-available plastic. (I added it to the MCC guide because it really is that good. Go make a friend with a Sam’s Club card.)

Things I wish we’d done differently:

  • More focus on Level 3 exploration. I think the scoring was right for high levels of play, but it was a little OP and a little scarce in our regional tournaments. If you were like us and captaining a quarterfinals alliance in blue bumpers, you were probably out of luck. Several of our kids were shrieking SUCCBOT! at Kickoff, which we tuned out because SuCtIoN cUpS dOn’T wOrK iN fRc. Well, surprise, they kinda did here. A grip-and-flip design like 3940’s was also pretty plausible and probably within our capabilities had we made it more of a priority.
  • While we made our hatch panel grabber from 95’s design, we never replicated the ability for it to extend outward for the rocket. That would’ve unlocked eight more scoring cycles (four hatches on the bottom of the rocket, then four cargo), which would’ve given us more flexibility in picking other cargo ship teams.

Overall though, 14-5 in quals with our highest seedings and first two alliance captain positions ever left me pretty satisfied with this season.

(Just to be clear this my personal opinion only and dose not reflect the views of my team or any other individual on my team other than my self)

I was on the fence when this came up, we did not want to do everything we did not try to do everything but time was invested into things that eventually were ether removed or over complicated the design this lead us to not being done in any reasonable amount of time(first to matches at first district defence with parts falling off) and many iterations and improvements during the season that could have easily been during build (mech and electrical was out of bag) major geometric design issues were discovered at dcmp. This all being true we did do what we set out every season to do is to be the to dominate in our niche area each year. But this year was different because 100% of team hours went to the robot or robot associated tasks (we did not punch a single button) lead to why I believe that this was a positive season but still was over complex and never saw everything we entailed it to be and what we had on in our first match of defence.

I would say that our 2018 was under reached to our team abilities and got us to priestly the same spot in the season (second pick of 8 at dcmp needing semis to make worlds )

If you want me to explain anything else @me an I will answer any of your respective questions.

This was us.

1 Like

I’d say we reached the right amount, and although there were many things like Center of Gravity and a less passive hatch intake that we really should have considered, I’m pleased with how the robot turned out

We almost overreached on an under frame dual pump, dual cup, suction climb that would’ve looked pretty magical before we pivoted to the switchblade.

7 Likes

The triple climb with a hatch panel passthrough was certainly an overreach.

-Mike

29 Likes

I kind of agree with this, although we developed our triple climb in a way that it didn’t drive any other part of the design. I’m very glad we went for our passthrough this season.

I suppose the triple was an overreach, especially considering we never got ours to work successfully in a match, but again, we set up the integration of them so that they didn’t effect the rest of the robot, and we always had it take a backseat to the rest of the robot development.

13 Likes

So I’ve recently switched teams, but I’m going to comment based on the team I was on during the 2019 season. I think it could go both ways. I feel like we overreached on our hatch mechanism. It worked very well, but it took much too long and was slightly over complicated. On the other hand, we should have added a cargo mechanism. We performed very well at the district level, usually being the primary scorer, but were very unprepared for worlds as a low hatch bot. There was time between events where we could have fairly easily added a cargo mechanism had we not taken the climber off for our 30 pounds. On the other hand, our goal was merely to make it to district champs and we qualified for worlds as 12th in the district, so maybe it wouldn’t have changed anything whether we had added a cargo mechanism.

Yea probably, building a Pink arm was a lot of fun, but a huge engineering challenge for us. It took away a bunch of focus on our manipulator design. Also something that I didn’t realize until later, we had to compromise a bunch on the acceleration of our drive base and how fast we move the arm to avoid the huge moving mass from tipping our robot and destroying itself. To no fault of our driver we just weren’t able to drive the robot as smoothly and effectively as years past because of the mechanics of the large moving mass that is a Pink arm with a big manipulator.

7 Likes