Offseason Testing - COTS 3 Motor Gearboxes and Wheels

After the first trip to Houston for our team members we are excited to start working on things in the off season to help improve our capabilities for next year. One thing we want to look at is drivetrain improvements. We have been running the KOP chassis each year but with changes to wheel type and gear ratio to help fit the game (have used 8" plaction and traction wheels as well as pneumatic on an offseason project). An idea for the offseason is to get a bunch of different wheel types and do some experimenting with them in both 6 and 8 wheel drive configurations to really see which ones work well for us in different scenarios. Build a few mock up game elements from the past (like the moat from Stronghold and bump from 2012) and test how well each configuration handles crossing them.

We have lots of old KOP chassis lying around so I would like to try to re-use them as much as possible to keep costs down. Some of the COTS wheel and gearbox options are geared towards a WCD chassis so I wanted to find out what success teams have had adapting them for use on a dead axle KOP setup. We also don’t have a lot of tooling so I’m curious how reasonable it would be for us to assemble some of these items (no mill, lathe, or CNC, and currently the most we have for a press is a 5" vise).

Some questions:

  • We want to try out a 3 motor gearbox so that it is an option for more acceleration and power if we need it, either the Evo Slim or WCP SS. The Evo Slim mounts easily but requires a 3 ton press for the pinion gears, is that out as an option without a press that size? In competition we would likely use Mini-CIMs so the CIMs you can buy with pre-installed pinion gears aren’t helpful.
  • VEX and WCP only list instructions on mounting the 3-CIM Ball Shifter to the KOP chassis, can the WCP SS (regular or flipped) be mounted the same way?
  • There have been reports of destroying the aluminum gears in the WCP SS gearboxes, where do teams get COTS steel hex gears to replace these? WCP and VEX only sell aluminum for most sizes.
  • Colsons require a hub to be pressed in for dead axle use. VEX has a hub but you can’t bolt all the way through the wheel to mount pulleys like with other wheel types we have used. Are the holes on the VEX colson dead axle hub tapped or do you have to drill through the wheel?
  • Does anyone sell a COTS dead axle 6" hub for pneumatic wheels? The only COTS hub I can find for wheels that small is WCP but it is for hex shaft. We have the 8" hubs for this from AndyMark, searching old threads it sounds like AM used to sell a 6" version but it has been discontinued.

Thanks for you input!

Easiest for us on 7214 is custom gearboxes I’m working on a design now that will mount to wcd chassis. Wheels we also will be using custom ones because of the huge demand and yearly low inventory we see from most suppliers. I can bring some of our stuff to State Champs, I know most teams don’t wanna hear it but going almost fully custom cut our costs by a lot plus we also control the timelines.

https://www.youtube.com/watch?v=uQ_mdx8_6A4

7 Likes

If you’re unable to modify the kitbot with much precision, it sounds like the Evo Slim isn’t a bad idea - you can find a nearby team that’s willing to let you use their press. The WCP SS issue can be mitigated with proper maintenance, as @maxxman stated.

If you’re looking for custom wheels, Andymark’s Plaction and Performance wheels are both fine choices - the Performance wheels especially are scarily light for metal wheels, and super robust. Colsons are good wheels, but using them isn’t going to magically improve your season.

Frankly, if you lack even a basic manual mill, it might be a good idea to go with a more default kitbot, and save your money for either robot mechanisms or getting more machining equipment. As cool as it is, the like, 5% more power that triple minicims gets you is just really not worth the cost or time for you guys, from what I can understand through this post. Focus on the parts of the robot that score you points :stuck_out_tongue:

Easiest for us on 7214 is custom gearboxes

As Donut said, 2662 has no precision machining equipment, so unless you’re expecting them to be able to lay out and drill with the 2-3 thou precision needed for a robust and efficient gearbox, custom is out of reach.

The WCP SS issue is not rare, district teams that run long seasons have this issue every year.

2 years of usage for us with 80+ official matches and we have had to replace before worlds, depends on how hard/long you run them.

Our practice robot this year went through 3 sets. We don’t view it as an issue, more of a general maintenance problem.

2 Likes

News to me, good to know! I’ll edit my original answer

To be entirely honest you can get competent power and acceleration out of the standard KoP gearboxes with 4 NEOs - and using either 6 CIMs or 6 NEOs is honestly overkill (I guess you can try 6 miniCIM if you’re insistent on trying out some COTS 3 motor gearboxes). Maybe adjust the current limiting proportional to your gearing and wheel size and its pretty much set and forget.
I used to be more on board with the CD trend of optimizing every little aspect of your drivetrain but after utilizing the kit chassis there’s really not much gained for going custom from a performance standpoint and your time is probably better spent researching more efficient and effective mechanism design. Especially with the introduction of NEOs, you can already get nice sprint acceleration without going for a more costly 6 motor setup.
And if you really want more traction you can just swap out the standard higrips with blue nitrile plaction - there’s really no other tread material that really can compete with the traction you can get out of the stuff.

I like the spirit of testing out different COTS setups and optimizing your drivetrain, but the differences in performance are more subtle than one would think and you’ll start to get diminishing returns despite spending hundreds of dollars on gearboxes and wheels. Of course, going custom is great from an educational standpoint, but before anything your team should invest in some precision machining capability first.

Really…we were giving him a third option that is usually cheaper we’re also another team in his state that can make gearboxes for his team.

Usually, this type of destruction is because of wear and tear. I had a time when they were assembled wrong and the teeth were all broken off. So be careful of that too.

No.

Pretty sure the VEX colsons don’t need hubs. Or have holes for you to drill through. You would just use the pulley on the shaft with the colson also on the shaft. Versawheels are a different story.
https://www.vexrobotics.com/colsonperforma.html
https://www.vexrobotics.com/versawheels.html

Do you want 6" hubs or 6" wheels?
If 6" pneumatic wheel, wcp is the only choice.
If 6" hubs, nope.

I’m going to split up my replies into a few posts to keep this from getting too long.

Glad to hear the performance wheels are good. We used 8" plaction wheels last year and they worked well (for those asking “why 8” last year?“, we wanted to make sure we could climb forks if there were buddy climbs). AndyMark has a sale on their plaction wheels right now and I grabbed a set of 4” ones at $13 each we can use.

I’ve wanted to try out Colsons because they have seemed to work really well for a low maintenance setup, but it sounds like they aren’t worth the trouble for dead axles. I also saw the notes from 254, 1323, and others after my post about Colsons having inconsistent tracking behavior for autonomous (not that we’ve ever had an autonomous close to theirs… but who knows in the future). If they aren’t going to be any better than a plaction wheel it would be wasted effort on our part.

For a live axle setup like WCD the VEX colsons already have integrated hubs, but they are designed for 1/2" hex shafts. For a dead axle setup like the KOP you have to get a standard colson and press in these:
https://www.vexrobotics.com/vexpro/motion/wheels-and-hubs/colsonhubs-g.html

Should have made this clearer, I am wondering about 6" pneumatic wheel options. We’ve used the 8" pneumatic wheels from AndyMark for a robot that drives in parades but their size limits what you can do with your frame for the KOP.

I didn’t emphasize it much in my first post but it sounds like wheel material and arrangement is the more important thing for us to experiment with. In the past our quick answer to any need to climb something was “use bigger wheels” but that comes with the disadvantage of more weight and higher CG. It also doesn’t always work; we still got beached in the moat every time we tried to cross it in 2016, and that was with 8" VEX traction wheels with roughtop tread. Being able to overcome field obstacles (or climb on top of partners) while still driving smoothly during a match is what I see as our objective.

For wheel material we just need to permanently move to something more durable than the HiGrips. We used straight KOP chassis this year but never drove off of HAB 2 after seeing the destroyed wheels others had from testing it. It’s better to not lose our drivetrain so we made the right choice, but those added 3 points every match for just using a better wheel type certainly make you an attractive partner.

For Deep Space and Stronghold, using pneumatic wheels over HiGrips would be the better choice. However, this doesn’t mean to permanently move to pneumatic wheels, as they have a whole slew of their own problems. However, we (3146) successfully used 8” Andymark pneumatics with the kitbot drivetrain this season but discovered that the kitbot did not have sufficient drop for pneumatic wheels, so we added a pair of 8” Vex omnis to the front. We moved off of the L2 hab at the start of most matches and the 8” omnis bore some brunt of the force but held up well all season. Team 133, who made Einstein in Detroit, utilized a similar drivetrain setup to great success. We also didn’t experience any particular mounting issues using the 8” pneumatics that you mentioned earlier as reasoning for looking into 6” options.

I actually saw plenty of teams that moved off of Level 2 with the standard white higrips and even one that “yeet” climbed up to Level 2 until the field staff told them to stop doing it as it was damaging the field. I visited their pit and checked out their wheels which held up from that aggressive manuever impressively well. So the HiGrips are also pretty durable wheels for the most part, as the design of their hubs have improved vastly over the past decade or so. However, with the no bag era of FRC approaching, I imagine competition robots for most teams would be also used as practice robots and hence experience more wear and tear. Both HiGrips and Plaction wheels might need replacements as over long periods of aggressive use they may develop cracks in the hub or the tread material may wear down. An Andymark employee or someone with more experience with these wheels can fill you in on the expected failure modes you might encounter if you put a lot (and I mean a LOT) more driving time on them. Of course this also extends beyond just wheels, but its something to keep in mind. This is actually one of the perks of using Colsons as the tread doesn’t wear as quickly and they’re just solid and bulletproof (Colsons have been a popular wheel choice in combat robotics as well, so yeah they are Durable). However, this isn’t that big of an issue as long as you make sure your wheels are fresh going into competition and to replace them when necessary before they fail (replacing them after a certain amount of aggressive driver practice hours or a rough competition) which doesn’t take that much effort.

The NEOs do open a new world of power with brushless, but they are a more expensive option in our case. Since we don’t have any a 4 NEO + Spark Max purchase would run $460 while I can get a pair of 3 motor gearboxes for ~$270. We have existing motor controllers we can re-use and a lot of unused Mini-CIMs from the KOP so the motor cost wouldn’t add much. I did think about playing with NEOs instead I am just a little wary of them being worth the trouble with all of the firmware updates required this year since our main programmer is graduating.

That kind of seals us not using WCP SS if it doesn’t adapt to the KOP chassis easily.

We were able to run 8" pneumatic wheels without problems, I was more noting that it inherently limits your frame options because of their size. Once you add in a few inches for space between wheels and the front & rear rail assemblies you effectively can’t have a robot shorter than 28". You also can’t do an 8WD setup unless you have a realllly long robot. Those are okay as long as you are willing to make those design tradeoffs; in some games it is nice to have the wide setup.

Glad to hear the Omnis held up well, that is the other thing to try out.

I hear you on this. My original idea was to get a bunch of wheels and a set of new gearboxes for testing as part of a specific grant in our area but I’ve backed off that and think we will just be trying 6WD and 8WD variants of Plaction, Omni, and Pneumatic wheels to see handling differences. The plaction wheels were on sale at AM as I noted a few posts above so we won’t need that much more to try out the rest of this.

Our students want to make changes to their scoring mechanisms from this year too, so that isn’t being forgotten about! Drivetrain is important though as we saw this year where many second picks were based purely on their defensive capability. Having something beyond the basic kitbot is important here if for no other reason than “winning” the tiebreaker when a scouting team thinks two robots have played similar defensively. There’s a reason that putting Mecanums on the DNP list is a meme.

On the machining note we would not have the expertise to use much more advanced tooling than we have right now even if we somehow acquired a nice mill or lathe. All of our technical mentors come from more of an electrical/programming background. My day job is more system integration which is why I like COTS so much, I’ve done basic mill work on a previous team but I don’t know proper machine maintenance and would likely destroy the tooling :neutral_face:

In terms of machining stuff, don’t let lack of experience stop you! You’ll destroy a lot of tooling, sure, but that’s part of learning :stuck_out_tongue:

Lots of great resources and everything online these days, I’d say it’s a totally feasible endeavor! I’d be happy to answer any questions you might have about that sort of thing.

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.