I’m a Chassis Lead member and I have no idea what I’m doing

Technical Technical Discussion
1

Hello all,

First time poster, long time lurker. I don’t know if this is the right place to post this so please feel free to remove it and direct me to the appropriate place. I’m a member of an FRC team, and am a student leader of my team’s chassis department. Everyone obviously expects me to know what I’m doing, but I’m still clueless after 2 years of FRC participation. So, I realize that these questions may be dumb or ignorant ones, or whatever, but:

What are the most important parts of the chassis and how do I suggest to improve each? So far we’re using the WCP drive and a very basic aluminum chassis.

If I were to do a design review with an adult mentor, what should I suggest about our chassis?

Again, first-timer, so I really do apologize if this post is really not conventional.

2

Hi. Congratulations on your leadership position!
That said, a couple of questions as well:

  1. Assuming WCP being West Coast Products, which “drive” of theirs were you referring to?
  2. What sort of prototyping do you have access to?
  3. Have you poked around at examples of other drive trains to base yours on?

Don’t worry about asking questions; asking around is how you learn more.

3

There’s no image of 6675’s robot on The Blue Alliance or CD-Media, so it’s hard to give you too much specific advice. But since it seems like you’re rolling your own custom chassis, you do have a benchmark to compare to: the AM14U3 that came in last year’s kit of parts. There are a few thousand of these running around each year either stock or mildly tuned. (And if you don’t have one in your shop, here are the documents on it.)

So take a good hard look and ask yourself: What does your chassis do better than the AM14U3? Can you swap wheels or do maintenance faster, package a mechanism or gearbox in a better way, mount bumpers faster, or is it stiffer or lighter or smaller or faster to assemble? If it’s better in every single way, then yay! If it lags in some area, is that area a big deal to your team?

Get a hold of last year’s driver and pit crew and see what they thought too. Was your robot fast enough? Were you getting pushed? Was it easy to control? Some of these things could be gearing or control system choices that aren’t strictly the chassis, but they are things to consider.

Now that you know the problems, you can start seeing where you need to improve. The answers might be to make small improvements on the current design, do some new custom setup as a fall project, or decide your team is better served by using whatever comes in the Kickoff Kit next year and focusing on other things. But you won’t be able to decide on a path until you’ve got a better understanding of where you are now.

(For the record, I do work at AndyMark that makes the AM14U3. But I would make the same advice either way.)

4

How much variation is your team possibly accepting? Is your team pretty much going to do a WCD for any game, or are they open to somewhat different chassis designs, and are you willing to consider them? (There are teams which do quite well with the “single drive” paradigm, including some repeating world champs, changing only wheels, gear ratios, and (when the rules indicate) number of motors each year.)

If your team is fairly locked in, learn all you can about the options which ARE under consideration. Download JVN’s spreadsheet and play with the drive train pages until you don’t have to think about how to enter the data, just what it is that you want to make happen. Do searches on CD and perhaps the wider world on the parts you typically use, and become familiar with the foibles and workarounds people have developed. In short, become an expert on your team’s drive system, so that you can smartly suggest appropriate wheels, gear ratios, and such.

If your team is more flexible, learn about the different drive systems which have been used in FRC, and their strengths and weaknesses. Find case studies of teams who have posted their engineering notebooks or build season notes or whatever they call it who’ve used these systems, so that you can intelligently guide your team to the correct choice. There are a variety of overviews of drive systems, mine (no claims it’s the best, but it’ll get you familiar with the popular drives and terms to search to learn more) is in the second and third documents here. Edit: Oh, yes, do the same with JVN’s spreadsheet in this case, too!

If your team is somewhere in between, do some of each!

And of course, remember what your team has been through the past few years. Looking stuff up is great, but if you aren’t familiar with what has and has not worked for your team, you aren’t going to be able to make your case (whatever it is) convincingly with your team. If your team has historical notebooks, read them (especially the chassis parts). If not, find some time to chat with the mentors about the history of your team’s chassis and how you got where you are now.

Another Edit: And +1 to CalTran and Billfred’s advice; intended as a supplement, not a replacement. We have used the AM14U2 the past three years (not much different than the U3) and have already purchased two U3 chassis which we’re 90+% likely to use next year. We’ve never built a KoP stock, but we’ve customized it 4 of our 6 years, and are quite happy with the results. In 2015 we pimped it into an H drive, in 2016 we had 10 wheels of 3 different types, overlapping in different planes, and last year we went 4 wheels, chopping off the front 3/8 of the chassis, and substituting U3 wheels.

5

A good start would be to study the “Strategic Design” video by Karthik. It will help you understand why the various people on this thread are recommending the things they are.

As a leader on an FRC team, you have to have more than technical skills. You need to be able to evaluate and manage risk as well as manage people and projects in a way that utilizes the resources available to your team in the most effective way.

6

If you are having trouble making a custom chassis a good but underrated (I believe) solution is to use the kit chassis. The chassis that comes in the kit of parts has had more than its fair share of engineering put into it. And if you really don’t want to use the stock kit chassis

7

+1 in any case, but in your role, ESPECIALLY if your team is open to change/variation in the chassis!

Again, +1: and note that Phil listed three distinct things, each important:

  • evaluate and manage risk
  • manage people [promote buy-in, keep them engaged and productive]
  • manage projects [to] utilize resouces…effective[ly]…

My amplification:

  • For managing risk, I really haven’t encountered anything more effective than evaluating each thing that can go wrong, and making 1-5 scores on each axis: a) the likelihood of encountering risk/failure and b) the impact/seriousness should this item fail. Plot the grid BEFORE starting the project, and fill in the unacceptable risks with red, then orange and yellow for the hard but acceptable risks, and green for those of little to no concern. The orange and yellow should fall on approximately diagonal stripes.
  • Managing people is way beyond this post. The key things are to understand what drives your people/what is important to them, and to present arguments and encouragement in terms that make sense to their mindset. Some people get this instinctively, others need Meyers-Briggs or other tools to help. Also, become something of a parent to the people under you; listen to them, give advice, and advocate for them up your team’s hierarchy. You may even wind up helping them move, or in a really rough year, move bodies.
  • Managing projects: scads available on-line. The key item here in my experience is to break up the tasks and the timeline into reasonably small chunks, set a schedule, and track progress in such a way that you know when you fall behind. The sooner you realize that you’ve fallen behind schedule, the more time you have to make adjustments to get back on track.
8

Hey it’s ok to not know what you’re doing. I’m glad that you felt comfortable enough to post on CD and honestly my best advice to you is to keep learning and never be afraid to ask questions.

One of the best parts about FRC is that we get the chance to learn really valuable lessons, technical and non-technical. Just do what you can to research and hold onto your curiousity.

9

Some great resources to check out: https://www.chiefdelphi.com/forums/showpost.php?p=1698030&postcount=13

10

On top of that, simbotics runs a few workshops on what they deem the “Kitbot on Steroids.”](http://www.simbotics.org/resources/kitbot) It’s a few modifications to the kitbot that make it significantly better than what you get out of the box.

Though the kitbot has changed throughout the years, watching these videos can give you an excellent understanding of why they decided to make those changes and how they made the changes themselves.

11

Unlike many other systems, Chassis can be one of the most difficult things to build, or one of the easiest. The choice is up to the team.

There are a couple interesting schools of thought worth bringing up.

#1: I will play around with different chassis in the fall, but would never choose a type until after the game is announced in order to ensure we have the right type for the game.

#2: Unless something crazy, or rule prohibiting, we will use chassis XYZ style, because we are familiar with it and it works well for pretty much any situation.

There are a lot of teams (including great teams) in both camps. The biggest thing I will say for chassis is being able to move is typically the most important single item for a robot. Because of this, be mindful to not bite off more than you can chew. If you think you are in over your head, choosing a complex custom solution will only make that worse, and likely lead to a frustrating season.

Some very general items to think about:

In modern FRC, 7 Feet per second is definitely on the slow end. Above 12 feet per second, you need to be careful about your electrical draws as you can have certain negative situations like brown-outs and breakers tripping.

There are a lot of great robots going faster than 12 feet per second, but most have 2 speed gearboxes. There are some great robots going faster than 16 feet per second on single speed, but they have to be very very careful with power management so as to not trip their main breaker or their circuit breakers during pushing matches.

In the meantime, you have a lot of time to learn about chassis before teh season starts. Being a young team, I would recommend learning about all the COTS solutions available to you. There are fantastic chassis and drivetrains from AndyMark, VexPro, and many other suppliers. To be frank, these COTS drives are better than some of the best custom drives just 10 years ago.

Learn how gear ratio and wheel size work together to give you ground speed. Learn about how smaller wheels contact the ground closer to the corners, but tend to be less forgiving on uneven terrain and typically wear out quicker.

12

If your team insists on a non-KOP chassis, see if you can create a rolling chassis pre-season. Not only will you learn how to apply some of the great suggestions above, but also what specific areas you need help with.