Is it worth building a custom chassis?

Hi, I’m from team 6884, and we are trying to decide whether or not to build and design a custom chassis or not. We are a 3rd year team, but we essentially started from scratch last year. We have never attempted to create our own chassis and have always used the stock kit chassis. Frankly, we are fairly lacking in experience and tooling.

Here are some pros and cons I have thought of:
Pros
-great learning experience
More design flexibility
-better performance

Cons
-time constraints
-lack of experience
-lack of resources/tooling
-I don’t see much benefit over the kitbot chassis considering we are an inexperienced team.

Is it worth it to build our own chassis, especially considering the relatively short amount of time until build season? Thanks in advance

P.S. thoughts on Kitbot on Steroids?

1 Like

If you have to ask, probably not. Give it a shot over the summer, but at this point, kitbot is a great choice.

15 Likes

The kitbot on steroids is quite outdated. As far as I am aware, nobody has posted an ideal KBOS for a long time for the latest gen of kitbot.

Your pros and cons list is pretty accurate, it would be beneficial for your team to reflect honestly on your skillset, resources, and a strengths and weaknesses analysis.

Good news is you’re asking all the right questions. In my personal opinion, stick to the kitbot. Focus on everything else, check out 118’s Everybot, and other Minimally Competitive robot ideas, like the one from Spectrum 3847 for example.

5 Likes

This all screams stick with the kit chassis. Trying to do it in-season is a bad idea (resources taken away from scoring and instead trying to figure out how to move when a perfectly good chassis is already given to put together). It may be good to learn in the offseason if you have the funds and manpower, but trying in-season is asking to not have a great season.

One of JVN’s blogposts about a team struggling with their custom chassis may be a good read.

7 Likes

Better performance than a Kit Bot only works if it really is better than the KOP. I assume you will build standard tank style? If then, it should be better but there are also a lot of stuff (chain/belt, in tube, 254 style, etc) that go into the process of making a good one. But if you choose to go with any other fancy drives, now is not the time to do so.

Tldr, lots of stuff go into the making of a good custom. There are places to go wrong, and the kop at this time, may just be your best bet.

If you want to try your hand at making a custom drive train definitely get something working preseason. Also don’t sleep on the kit bot. In 2017 my team used the kit drive base with a VEX shifting gearbox after many years of custom drive trains and had a breakout robot performance year. Using the kit drive-base saved us so much design and build time and allowed our programmers and drivers to get their hands on the robot earlier.

1 Like

It depends on the game.

My daughter’s last year, at the first meeting after kickoff, some of the veteran team members went off to discuss strategy. The rookie and less experienced team members were asked to put the kitbot together as a training exercise.

After they got it finished, the stratigists looked at it and went, “Hmm. Hmmmmm. Looks like that will work. OK, we have our robot.”

2 Likes

I think you have a good grasp on the pros/cons. Just keep in mind that the time you spend on a custom chassis is going to come at the expense of something else.

I think very few teams are primarily held back by the KOP not being good enough to them, compared to not having good enough manipulators, not having enough driver practice, or not having enough programming time. I don’t know your team, but I think it’s important to ask if this is the best use of your time compared to the alternatives.

5 Likes

@Ryan_Dognaux, I believe you have something to share here? :slight_smile:

Last year the then 4th year team I mentor decided, to my surprise, that they wanted to build an H-drive. They came to this decision about a week into build season.

Perhaps this is approximately your situation. So, the Pros and Cons of how it turned out:

Pro: They did it. Albeit only with a modification of the kitbot system but still it was quite clever and worked well. They climbed straight up that steep learning curve.

Cons: It did detract from time that could have been better spent on manipulator systems, although the apocalyptic weather we had in Wisconsin last year was a bigger factor.

To some extent this sort of question comes down to what kind of team you have and what kind of goals. If learning things and creating cool stuff is set higher than winning blue banners…go for it.

But 5826 is a peculiar bunch.

T. Wolter

+1 for all of these - they were my thoughts before I started reading the thread, you ya’ll got there first :slight_smile: .

In my own recollection - 2016 was a year we built a custom frame and chassis that I do truly believe was more effective than the kitbot. Most of our designs are based on the lessons learned that year. However, prep for it was about three months of a personal project for a mentor, and another two with the team building and experimenting. We’re usually pretty slow at doing stuff, but it’s at least another sample point to say “do the hard part over the summer”

We did this in 2015. Of all things on the bot, it worked pretty well. Wouldn’t necessarily recommend it again though, there are often better ways to get side-to-side travel.

Let us know how your choice works out!

1 Like

Similar situation here, we started from scratch 2 years ago and were in your position last year…
Agree with everything you said but would like to add one consideration.
There are modifications to the KOP chassis that do not require significant engineering work that you might consider based on game design .
Pneumatic wheels can be handy and are not too difficult to add. Not cheap though.
You can also easily and cheaply modify the gear ratio in the Toughbox Mini or (more expensively) add a COTS gearbox to the KOP chassis.
And you can replace the CIMs with brushless motors.
Last year we kept the KOP chassis and put all our efforts into building our first elevator and wheeled intake. A custom chassis would have been a mistake.
We designed and built a WCD over the summer which was educational on many levels (lots of mistakes…I mean learning opportunities). CAD, CAM and machining never goes how you think it will the first time. We have been designing different versions of the WCD (number of wheels, size of wheels) this fall. We opted out of the KOP for the first time.
If you have not opted out, That is another reason to stick with the KOP.
This is just our experience and is backed by little more than that. Hope it helps somehow.
Good luck.

1 Like

Oh boy do I.

The KOP chassis is the best thing you can do to have a successful season. Your team can have a drive base up and running within a few short meetings with absolutely no experience. Brand new team members can use the step by step assembly instructions to put it together. It is built robustly and will not fail you.

There are also some nice, simple upgrades that you can do if you have the funds. Adding brushless motors like the NEOs will improve performance and AndyMark sells other toughbox ratios if you need to go faster or slower depending on your wheel size choice.

The resource points you’d spend on a custom chassis can then be spent on the really important part of build season - game piece manipulation.

Choosing the kit chassis last year was easily the best decision we made. Our drive train performed flawlessly through all 75 matches that we played. You don’t have to build it exactly to the out of the box configuration either. We went with 8" pneumatic tires to come off / on level 2 easily but still used the stock belts and pulleys. We swapped for the largest spread in the toughbox and upgraded to 6 NEOs for IRI. Pushing other robots around was not an issue at that point.

image

image

In summary, the AndyMark KOP chassis is probably the best thing to happen to FRC ever and you should use it.

Full disclosure: I’m the drive coach for team 1720, we ended up 3rd in Indiana last year and I love the KOP chassis.

14 Likes

According to Mike Corsetto, technical mentor from 1678 you should use the kitbot, you should spend your time focusing on the mechanisms and not on the drivetrain

Watch 1678’s fall workshops, sepcifically the strategic design video

100% agree this close to build season, go with the kit bot.

After your season starts winding down have the students get some basic CAD practice whether SW, Inventor, F360 or OS (all discussed in depth on these forums)

Then look through these videos by Yeti for a great summer/fall project

Seeing that picture of your robot reminded me, you need to poke a hole in the end rails of the KOP chassis, so you can easily install the nuts that hold the frame together.

frame

Full disclosure: I’m the drive coach for team 1726, we ended up 2nd in Colorado last year and I love the KOP chassis.

10 Likes

Wow that’s a big hole! A student could probably fit their whole hand in there :laughing:

We actually ran the Gen 3 frame last year so it was a little different. But I highly recommend getting the frame bolts / nuts all put in and loosely tightened before putting the wheels in. We’ve done some goofy things but usually a long socket extension can fit on there enough to hold the locknut in place while someone else tightens the head of the bolt.

With the Versaframe system, you can basically get a hole saw and use that to make a drivetrain. If you have access to a good CNC machine, you can even cut your rails on there. 6502 has been doing custom west coast for two years now, and haven’t had any real problems with time after getting comfortable with the design. I would recommend building one in the offseason for practice so you get the hang of it though. Good luck!

The corollary to this is “I have seen quite a few teams primarily held back by spending too much time and resources to build a custom chassis that performs poorly (or no better than the KOP) and not having good enough manipulators, not having enough driver practice, or not having enough programming time”

2 Likes

Great tip, Jim.

Hope to see you and your team here in Houston again.