I am part of a team whose rookie year was in 2019, and we are relatively inexperienced but have enough knowledge to get by. I was researching the possibility of mecanum wheels on our bot this year, and after reading over the AndyMark guides it seems very doable. We have a great coach and a lot of hard workers, but I wanted to ask if anyone out there had warnings, advice, or other information about mecanum wheels.
- Jack, 8230
Running mecanum wheels will make you very susceptible to defense. They are very easy to push around due to the rollers and using them can make your team a target for defense.
If you really do want to run them, you will also have to watch out for CG. Offcenter cg can cause weird strafing behaviors.
My advice would be: don’t. A well done tank is nearly as good as a swerve, without the numerous drawbacks of mecanum.
From a fundamental standpoint it is almost never advisable to try a new drive train during build season as it’s probably the easiest way to ruin a competition season if it fails. As others have said a mecanum drive train leaves you with no traction to push other robots or fight back against robots pushing you. In a game with a wide open field this is something you want to avoid. On a more positive note, a mecanum drive train is a perfect off-season project.
Agree. But with correct calculations of what gearing you use on each of the four gearboxes can make you very effective at avoiding defenders. Bad calculations will make you Bambi on ice.
Also agree. You don’t want to have great mechanisms on a chassis you can’t figure out.
That said… in 2017 team 1986 did pretty darn good with a mecanum system. They probably didn’t try it for the first time during build season.
I want to believe. I really do. Unfortunately, I’ve never seen this happen IRL.
The closest I’ve seen was a team at a FIM district in 2017 who used 2 mini-cims on each gearbox. They were hella fast and even managed to mount some defense of their own—until they drove into an airship at full speed and never moved again.
We did mecanum in our 3rd year and thought we were pretty slick until we made it to state and got bullied by every team we met. It was like someone put out a memo in the pits to push us around.
In 2019, 2539 ran mecanum as a pretty experienced team and did quite well. Ranking Number One at both FMA district events and 4th alliance captain at world championship in Michigan on the Archimedes field. The robot was great, mecanum was great but I would still not recommend it. We actually switched the robot to tank for the off-season and I would say the robot moved well enough. The big difference was how other teams reacted to mecanum. We had to be an alliance captain at each event, otherwise we would not be picked for finals regardless of our performance during qualifying. It didn’t matter that we could solo a rocket or climb hab three everytime or had fast cycle times. Most other teams seem to only see mecanum as weak against defense. I would argue that we did stand up very well against defense with a strong driver and smart use of safe zones. But there is an overwhelming consensus against it and I would focus on other aspects and mechanisms of the game.
Btw, we run swerve now.
Actually, we did.
I have been on two different teams that used mecanum drive trains. While it is possible to get your calculations correct as @SenorZ stated, there are a whole bunch of other factors that can trip you up and ruin your season. For instance, one team had a very rigid chassis and it was warped slightly so one of the mecanum wheels had less pressure on it than the others. This made it drive in an unpredictable manner. Neither team actually used the mecanum drive to go sideways in their competitions because their drivers were not accustomed to driving like that so they would drive the robots like a normal tank drive robot.
Gonna preempt this by agreeing with everyone else, Mecanum isint worth it.
Theyre rare, but they exist. If you want a pure mecanum, i highly recommend looking at 492’s 2017 season. Despite being the #1 seed and finalists in Galileo, they seem to slip under the radar for one of the few good implementations of a mecanum bot.
As you probably gathered by now, the advice is “don’t”, the warnings are “It’s a trap!”, and the information is “do it in the offseason first”.
If you DO do mecanum wheels, the key thing that will help you is to get the frame put together and the wheels on as soon as possible , preferably yesterday or the day before, so that you can practice driving. Put some weights on the robot to get it up to 150 with battery and bumpers on and drive, drive, drive, change batteries and drive, drive, drive some more. Play around a bit with the CG for maximum effect.
As a team that is notorious for making a lot of mistakes, trying to build a new drivetrain during the season is one we will never repeat. Ever.
Last season we invested in SDS MK3 system and it is awesome. My students were able to put these together at the same time we were starting to learn how to run our new CNC router - and the build was simple and straightforward. We (or programming/mentor) has been smoothing out the wrinkles.
So we have a working Swerve drive. And though I wouldn’t call it smooth, it is functional. There is 0% chance we use it this year.
The lesson? Use tried and true easily adapted drive trains. The drivetrain is the most important part of your robot. Build one that is dependable and suits your programmers levels.
When would you feel comfortable implementing swerve on a competition robot? It sounds like you have a functional swerve drive and CIS is no novice team, so why not this year? This is making me wonder if we are too rash with our decisions…
My guess is that they have them functional, but not smooth–and if there’s one thing you need with a swerve drive, it’s smoothness.
Without being CIS, I would guess the earliest they’d go would be 2023 season or 2022 offseasons after they’ve got the bugs worked out and gotten some drive practice.
Or maybe never… 330 once spent two offseasons working on a mecanum drive. 14 seasons later they shut down and they’d still never used a mecanum drivetrain on the field.
I completely agree with the advice against trying out a new drivetrain in competition season.
That being said, however, beware of the strong Chief Delphi bias against mecanum. The team I was on in high school (418) ran mecanums in 2010, and dealt perfectly well with the massive (literal) speed bumps on the field that year. That same robot also maneuvered just fine, holonomically, in my bumpy, muddy front yard while I was “transporting” it for a weekend demo event. (Sorry, @Danny_Diaz and @Bertman; if I’m going to have a robot at home overnight, I can’t resist the temptation to drive it a little.)
The point being, mecanum definitely has its advantages and disadvantages that you have to balance, just like any other engineering trade. But if you hear people on CD saying “mecanum only works on a perfectly smooth field” or “mecanum is useless against defense”, that’s just hyperbole.
We used mecanum on our 2020-2021 bot. After reading the posts above, perhaps we lucked out. We got the gear ratio correct. We used it to strafe, and it was excellent (we had a power issue with our switch so we were stuck in a few matches), but the drivetrain worked well.
We did not have too many issues with defense against us. The limelight helped.
If you go this route, one of our programmers/drivers had a good solution to force drivers to use the robot well.
We used a game cobtroller, and they mapped forward and back to left stick Y, and strafe right/left to left stick x. Turn was x on the right stick (turn to target was mapped to a button as was our shooter speed).
This control setup helped a lot.
There was a huge sacrifice in playing defense though.
We played defense some, and the drivers were incredible. Yet, the robot could be pushed out of the way much of the time.
It did not matter much in such a tight feild. We determined that this year is different though.
Please, never gauge your decisions based on what 4607 is doing. Most times we don’t take our own advice. But if you really want to waste time with a study on consistency, my team is your huckleberry.
We kinda did try a new drive train in a build season (2016/'17 Butterslide) - I cannot even look it up to infuse a hyperlink; it was a disaster that sank a lot of resources into a drivetrain that didn’t stratch any one itch. At Champs in St Louis we tore the whole thing down and bagged a bunch of hardware and components. In 2017 logic - we were supposed to ship a robot back to MN in a bag so we could compete in the MSHSL State Tournament. We shipped back a small hardware store. We rebuilt the entire robot to be a gear runner for the MSHSL tournament. It sucked.
Tldr; Building a new drive train in build season was expensive, complicated, and sucked.
Be rash. Live on the edge. Especially this year where it is hard to gauge expectations and the field is wide-open. If not this year, then when?
We will use our swerve when we feel ready. That doesn’t mean in a FRC game - it could be for some random task, a cool demo bot, or something that we utilize when we can afford the SDS MK4 (honestly - a pretty big reason at the moment).
Do what your team is wanting to do - but also tether that to what your team’s capabilities are at right now.
Running mecanum is sort of an all or nothing proposition. You either need to be good enough to be a primary scoring robot on an eliminations alliance or bad enough that it doesn’t make sense to be playing defense against you.
There are at least two different ways to evaluate this choice:
- From a competitive standpoint
- From a fun standpoint
From a competitive standpoint, this is unlikely to pay off. And if you really need the movement I would recommend holonomic instead.
From a fun standpoint: I remember having a fairly heart-wrenching discussion with a member of a different team where he wanted to know where I thought his team would get picked. His team’s robot wasn’t bad - realistically number 20-24 at the event in terms of offense - but there was no way that they were going to get picked because at that point in the picking teams were going to be looking for a robot with at least the option to play defense.
I’m curious what you think the definition of “holonomic” is.