One of the shafts of the wheel stuck out during the match, this resulted in only one side of our robot available moving. We want to fix and upgrade the robot.
How to decide which kind of wheels to use? Omni, duraomni, mecanum and pneumatic? And what size is better? 4" or 8"?
It depends on what type of bot you want to have. If youâre a primary defense bot, donât go omni or mecanum but if you want to be very maneuverable omni and mecanum can be good choices
Then likely the trouble is not with your wheel choice. A normal KOP drivetrain shouldnât have issues. It could be motor controllers, wiring, or programming. If an axle bolt is sticking out, look for missing spacers.
Having an axle bolt âstick outâ and locking one side of the drive train on a recent KoP chassis is pretty strange. The plate separation is held pretty constant by a combination of the ties into the end sheets, churros, and those 3x1 top plates, and the shafts are 3/8" bolts with nylock nuts. Once you get those in place, they should be pretty solid under any reasonable amount of misuse. Do you have a picture?
Also, if your robot was performing well before the shaft issue, I wouldnât change the wheel size unless you compensate with a different gear ratio. If it was accelerating sluggishly or you were drawing too much current, smaller wheels (or a lower* gear ratio and the same wheels) may help with this. If you could accelerate really quickly, but couldnât get enough top speed, larger wheels (or a higher* gear ratio and the same wheels) may help with this.
As noted above, changing among solid, pneumatic, omni or mecanum wheels is entirely a function of what strategies and tasks you are taking on with your robot.
* A lower gear ratio is one with a higher number, and vice versa - that is, 12:1 is lower gearing than 9:1.
If youâre going to be competing replace the wheel with whatever type you had on to begin with (your drivers being familiar with how it drives is probably more helpful than any theoretical improvement gained from changing it) and spend the time trying to look for why it broke in the first place, The KoP chassis really shouldnât be breaking.
If youâre in the post-season then try everything! Experimenting and seeing what works and what doesnât is half the fun of engineering.
Proper design and meticulous craftsmanship â required for a mecanum bot you wonât regret â can be challenging, but what part of programming mecanum did you find challenging ?
Typically if itâs a flat field (2017, 14, etc.) Its in your best interest to use the smallest wheels possible to keep the CoG low. In cases like this, 4" wheels are the way to go I feel like. Good balance when doing a WCD style of keeping everything low but still a reasonable ground clearance (roughly 3/4" depending on belly pan. Colsons are a favorite of mine since they are very durable and our set in 2017 lasted us all season (2 districts, state, champs including Einsteins, IRI and WMRI). I know teams in the past have had consistency issues with colsons (254 and 2767 are two that come to mind) but it just depends how you are controlling your autos. When using colsons, omnis do help with turning but arent a necessity. It does seem to help with consistency, making sure where your rotation point is.
Traction/pneumatic - way more grip, youâll probably win pushing matches more, but turning is harder. Autonomous will be less accurate. Drop center will likley be needed
all-Omni - less grip, side-to-side slippy. Can be more agile and turn in circles faster. Silly unless youâre doing an H drive
4-omni, 2-traction - a bit of both - a decent amount of traction, but not as much as all-traction. still fairly agile. Pick which pair are traction to control your center of rotationâs position (can be useful in some designs).
Larger wheels help you get over bigger bumps, smaller ones are often easier to design with if you donât need to go over bumps or ramps. Wheel size choice affects gearing choices to achieve the same top speed.
It sounds like you lost your e-clip on your toughbox output shaft maybe? Hard to tell from your first message. Make sure this didnât fall off of your output shaft -
If thatâs the case then your output shaft could slide out of the gearbox and it wouldnât be very fun to put back in. Youâd need to remove the toughbox to replace it, I think.
The three most common side moving drivetrains are:
Swerve, which you can tell typically by the pair of vertical motors at the corners
Mechanum, which you can tell by the diagonal rollers on the wheels and also the horizontal motors at each wheel
H-drive, which has 4 omni wheels at the corners as well as one omni wheel in the center âHâ position. This allows the center wheel to move the robot side to side.
On the topic of wheel selection, I tend to look at several factors when selecting my wheels.
How much ground clearance do I need to drive over field objects? Select a wheel with the appropriate radius to support this. Knowing how the wheels interface with your frame impacts this. A flatter field enables use of smaller wheels (although my personal preference is to not go beneath a 4" diameter wheel in a tank drive).
Whatâs the minimum gap between wheels required to avoid bottoming out while driving over field obstacles? If you know your frame length/width and the wheel radius selected in 1, you should be able to model this in CAD. We use this to determine the quantity and spacing of our wheels.
Does the terrain of the field mandate certain traction characteristics? I will select the âtreadâ of my wheel based on what types of surfaces the robot will be driving on, if there are inclines to drive up, and which wheels will be in contact with the playing surface at what moment. Games that have ramps or bumps to climb will generally point me towards using all âtractionâ wheels, rather than an omni on one end.
Do I want specific turning characteristics? This will impact my choice of center drop and/or omni wheels. You can manipulate how your robot turns and where it turns around by adjusting the relative traction of each wheel. Knowing your wheel quantity, wheel placement, track width, and frame length are also important here.
Weâve had enough consistency issues that we now print custom hubs and wheels and vacuum mold our own treads. We were able to cut the error in ticks per inch from about 6%+ down to 2.5%. The biggest issue we faced during FIRST Powerup was the unpredictability of the carpet on the swerve.
@rkalnins To be clear, you were using a smaller diameter Colson, correct? I donât remember the exact size of the wheels on your swerve, but theyâre significantly smaller than those used on most tank drives.
I remember having this conversation with @Jared_Russell after 254 posted about these issues in 2018, and they were using 4" Colsons. While I donât have any recorded data like you presented, Iâve never observed this with 6" Colsons, so I hypothesize the smaller diameter wheels may have some impact on the contact patch with the carpet and thus the deltas your teams experienced.