Pros and Cons of a West Coast Drive and Kitbot chassis

Hello all! I’m investigating the pros and cons of different drivetrains, namely a custom west coast drive and the kitbot. This summer we (6672 Fusion Corps) have been building our cad/design capabilities in anticipation of the arrival of our new Velox 5050 that will come in the next month or two. I believe the CNC will revolutionize how our team works since before we mostly hand-drafted mechanisms and made up the details as we went, and in the future we will be designing much more of the mechanisms before they are produced. This brings with it a lot more capabilities in producing much better and less janky robots.

First, some context to our situation. In the last three years, we have built good robots that have performed well but not always great. (we’ve had our highlights like ranking number 1 at our rookie regional, winning a district event, and being a captain at worlds this year). However we’ve usually had problems that have plagued our season. In particular, last year our drivetrain suffered horrendously. We were browning out our whole first competition due to a huge amount of scrub from our pneumatic tires and an under-powered gear ratio, with no good way to fix it because of it being the kitbot. We fixed it but the solution was to slow the acceleration of the robot a huge amount. Overall, the drivetrain was what was holding us back last year.

One of the ways we have been exploring pushing our robots to the next level is switching our drivetrain to a west coast drive from the kit bot. An alumni of the team and I designed a WCD over the summer and want to test it out in anticipation of possibly using it during next season. However, the mentors are not totally on board with using it during the season. To help us make a good decision, I am going to go through the Pros and Cons of each and I would like to hear what y’all think.

West Coast Drive:
Pros:

  • Good learning experience and we have the opportunity to make revisions between the fall prototype and the season robot
  • Easier to mount mechanisms than the kitbot
  • Easier to work with in CAD because you don’t have to worry about wacky imported files
  • We made a fully customizeable drivetrain assembly where all you have to input is the variables of the drivetrain and it outputs a full CAD of the drivetrain - so not much time will have to be spent on making a new drivetrain every year (just changing variables to match the game)
  • better performance and more flexibility when compared to the kitbot
  • I think if we opt out of the kitbot we can use the credits to get other stuff? I’ve been told that but couldn’t find it anywhere on the internet
  • more flexibility on gearbox choices (flipped 3 neo, two speed 3 neo etc)
  • utilizes our huge investment in the CNC, paves the way to more custom mechanisms

Cons:

  • More maintenance when compared to the kitbot (chain tensioning etc) and we will have to be more vigilant to making sure everything is okay with the drivetrain
  • Mentors think it will be very hard and are weary of it already; additionally they think that when I and a senior have graduated the team will not be able to continue with the WCD (I disagree)
  • It is not extremely simple like the kitbot
  • requires full CAD capabilities on the team, something that we are just now building

Kitbot
Pros:

  • tried and tested, which the mentors like
  • Extremely easy to implement, doesn’t take much to get it going as everything is already designed
  • We can get it easily and don’t have to cut tube and gussets during the season
  • Mostly reliable other than a bad season with it
  • very robust

Cons:

  • Not very flexible for things like drivetrain gearboxes in the back/ front or custom/flipped gearboxes
  • harder to work with in CAD
  • no room to grow
  • Harder to mount mechanisms to it (at least less flexibility)
  • It did not live up to what we needed last year
  • less wheel choices and less custom gearbox opportunites

Also, I was reading the Limelight documentation to try to understand as a build team leader what the programmers need for reliable vision on a robot, and I noticed that the limelight seems to need a stable base (I’m thinking no center drop on the drivetrain) to be most reliable. Does a center drop throw off vision systems when calculating distance to the target? Is it noticeable?

Final word: I want to implement a west coast drive on the team for several reasons:

  1. I think it will increase our capabilities for custom mechanisms and will utilize the new cnc (my thought is "why do we have the cnc if we aren’t going to use it for game changing things like this)
  2. I think it will be a good learning experience that can be continued into the future
  3. we will have greater flexibility in drivetrain design and a higher performance from the drivetrain when compared to the Kitbot
  4. It is cool to have a custom drivetrain
  5. I think our team has reached a level of maturity (good amount of students, a lot of support from our school and school alumni, adequate budget, good machining capabilities etc) that will allow us to push our robots to the next level and I think this will be a good first step.

Thanks for reading to those who did… sorry that this was really long :slight_smile: What do y’all think? Is it worth it if we can get a prototype working in the fall? should we still go for the Kitbot? I’d love to hear any thoughts.

I think your topic looks like this topic.

Build one, test it, see if it is worth the expense and hassle. If you can’t get one built out and tested in the couple months before the opt-out deadline, you should probably stick with the Kickoff kit drivetrain.

1 Like

I don’t agree with all of your pros and cons. I am a big fan of the recent kit chassis, and in particular I find it easy to attach superstructure to it, especially VersaFrame infrastructure. Start with some VF channel bolted or riveted down to the top of the chassis (you’ll want to bore some holes out), and build up from there.

Still, at the end of the day, there are teams who thrive with the WCD. I fully endorse your plan to do some fall testing of a robot built on a WCD base*, then proceed with 2020 based on what you learn this fall.

* If at all possible, design this robot to play DDS, and bring it to an offseason event to get a more realistic evaluation of how successful it is. Also, (unless you rebuild your competition season robot), keep in mind that this robot was designed post season with better intel on how the game was actually played.

1 Like

We made this move 2 years ago. We found the WCD maintenance to be faster and easier. Without the outer plates on your chassis you have much more access. We bought the parts around the month of September and used it that following year on our competition bot. What we did is built it in October then disassembled it and that way we were a bit familiar come January. Like others have mentioned, I think you should try it out and if you feel confident come January then do it, if not stick to the kit until you are confident.

4 Likes

When moving to a new machine, it’s nice to built your aux mechanisms around it. However, when you are just getting started and haven’t super dialed in the ins and outs of the machine, I’d be very hesitant to build a drivetrain (your absolute most critical mechanism) relying on it.

Kitbot

Some assorted thoughts on this topic.

  1. Continue to develop your CAD skills as a team, regardless of which choice you make. It’s a worthwhile investment, and essentially a pre-requisite to getting any real value out of your CNC router.

  2. Learn and develop your skills with your CNC router, regardless of your drivetrain choice. But depending on how quickly you pick things up in the fall, don’t necessarily rest your entire season’s success on using it frequently for 2020.

  3. WCD don’t require a ton of CAD (and even less CNC). While some drafting and planning are required ahead of time, it’s fully possible to build a WCD with some basic dimensional sketches and bench tools. My team has a video on layout of a WCD rail without precision machining.

  4. If you pursue WCD, I would highly suggest using VersaBlocks at the bare minimum. Possibly the Versa gussets and Versa Frame as well, depending on your CNC confidence (gussets being more instrumental than the frame). The VersaBlocks save a lot of time and effort in the bearing mounts, and make chain tensioning quick and easy.

  5. The Toughbox Mini gearbox that comes with the KoP drivetrain has five ratio options to pick from.

  6. There are numerous options and ugprades for the AM14U4 Kitbot Chassis, including the EVO gearbox family that provide numerous ratio options. There are also ways to mount VEX components to kitbots. You may not have run out of mileage on the kitbot chassis just yet.

1 Like

Here are my current impressions of your situation.

Identify the Problem
First, I would like point out that the issues with your drivetrain would have occurred whether or not you used a kitbot or a WCD. The pneumatic wheels would have still cause too much scrub, and the gear ratio would have still been under powered. It sounds like consideration of wheel type, gear ratio, and a couple other factors would have been more important.
I wouldn’t deny that increasing serviceability wouldn’t have help to make those changes, but it important to consider which factors really affected your drive.

Kitbot Credits
Every year, my team opts out of the kitbot and uses the credits for something else. It’s more cost effective to buy what we need for a kitbot, and get the credits than have AndyMark sent us duplicate parts we don’t need.

Should I switch?
For any team looking to figure out if switching drivetrains, remember that you should always build a drivetrain in off-season and test beforehand whenever possible. This will make sure you discover the hiccups beforehand.
However, this takes time. Could this time trying to optimize the mechanics of the robot be better spend upgrading the programming and electronics? Can you optimize you own drivetrain by testing wheel combinations and different gear ratios rather than switching? Always look for what change will make the biggest impact. Once you achieve that, consider the upgrade.

2 Likes

Note that this issue is essentially independent of WCD vs KoP. If you need the drivetrain to be stable when using the targeting, the easiest way is to move your CoG backwards or forwards, so that the drivetrain is on the same 4 wheels except in significant acceleration (CoG forward) or deceleration (CoG to back). If you really want your CoG over the center, go to an 8 wheel drive train, or use omni wheels on some corners so you can eliminate the drop center. (This last bit IS easier to do on WCD than KoP.) And never expect a stable drive train with pneumatic wheels.

Here is my take on the issue. I was a student in a team that swears by a custom west coast drive and am now a mentor in a team that will be using kit bot for the foreseeable future. Kit bot is easy, tested, and will probably get 90 percent of the performance of a custom west coast drive. Most of the drivetrains that preform sub-optimally are custom drivetrains that ended up being worse than the choice of kit bot. That being said I say that you should make it for the off-season and see how that goes. Now let me address some pieces of your argument between the two that I don’t necessarily agree with.

-More maintenance: I don’t think that my former team in 4 years of running WCD ever had to do maintenance more than greasing gearboxes every competition. Chain stretch is a non issue unless you’re building incredibly complex autonomous routines or are using the same robot for driver practice and competition in which case there will be exponentially more wear.

-Dropped center: I can not emphasize this enough, drop the center wheel(s). Vision is not that picky and having run vision systems all 4 years the drop of a drop center is not dramatic enough to throw off a vision system. Not having a drop center essentially doubles the length of the wheel base and will make the robot extremely scrubby and generally an awful experience to drive.

In closing I think that the big reason why teams do custom drive trains is to easily build 2 robots. Having CNC machinery is what opens the door to making 2 robots affordable and even with the changes to bag rules having a separate robot for the programmers and a separate robot for mechanism iteration greatly streamlines workflow.

2 Likes

One other thing to be concerned about. Just because your team has a new piece of equipment doesn’t mean that all your worries are over. Last year our team had access to 2 CnC machines and a 3D printer in our shop.

Thinking we could go from CAD to robot was a mistake because we found out we didn’t have enough time to make all parts (multiple sets) in our shop and had to have a lot of parts done by sponsor’s machines. We meet for 2.5 hours after school and found that many parts took over 4 hours to complete. We also did a off season project to test the new CnC machines, we just didn’t test the production rate.

We had 2 milling machines, a lathe and a student that could weld that went underused. I guess we also needed an industrial engineer as a mentor.

1 Like

As a little bit of an aside: yes, the Velox will probably revolutionize the way that you make parts however it will also have a learning curve. You’re going to break some bits and screw up some parts. Don’t plan to make 100% of your robot on the Velox if you don’t have some experience with it yet.

If you get the Velox far enough in advance of build season (which it sounds like you will), then I’d recommend making a WCD as an off-season project. Figure out what the problems are, how much work it is, and whether you think it’s worth it. That way you can have both options on-hand and make properly informed decisions about which one suits your team the best, and you can enter the season with some experience making parts on the router.

4 Likes

This is exactly what my team did for the 2018 off-season with our new Omio CNC. It was an amazing training exercise for our team and allowed us to turn out a modified version of that WCD drivetrain for 2019 without breaking a sweat. 10/10 would recommend!

If you have questions about our experiences, send me a message!

1 Like

We ran a WCD and had no problems whatsoever after we finalized the chain tension and had everything tested before stop build day.