How to make your robot withstand the beating

[RoeeVulcan]

This year me and my team build a pretty strong robot, but it kept getting problems with a few mechanisms that were not built to be so strong, one little match were we had a comunication problem almost disabled our robot for the entire day (take into account it was the first match of the day).

So I was wondering, what do you pro teams do to build stronger robots, that can stand the thrill of the fight?

[MrForbes]

Build several that break. Each one will teach you many things not to do. It's called experience.

If you have the opportunity to spend some time looking at some of the really good robots up close you can learn a lot. Ask questions about why parts are the way they are.

[RoeeVulcan]

Originally Posted by MrForbes

Build several that break. Each one will teach you many things not to do. It's called experience.

If you have the opportunity to spend some time looking at some of the really good robots up close you can learn a lot. Ask questions about why parts are the way they are.

I opened this thread in an attempt to learn from other teams experience, guessing they had a lot to teach me. I would love to know what your team did this year to a assure themselves on a robot that would come out of the field the same way it got in

[Daniel_LaFleur]

Originally Posted by RoeeVulcan

I opened this thread in an attempt to learn from other teams experience, guessing they had a lot to teach me. I would love to know what your team did this year to a assure themselves on a robot that would come out of the field the same way it got in

Ok, I'll give you a few tips. Depending on the game and your robot design, they may or may not apply though.

1> The KoP frame is very strong in the x and y directions but is flexible in the z (specifically twisting in pitch and roll). This could allow the frame to permanently deform with the 'vigorous interaction' with other robots (or this year ... falling). We usually box our frame. That strengthens it (and makes it heavier) and gives us a second deck to attach devices.

2> If you don't have to cantilever your drive system, don't. If you do, then calculate the twisting forces (especially side loads) and make sure your gearbox mounts can handle that.

3> If you use the c-channel, wherever you attach something put a wood block (or other support) into the c-channel. I've seen too many crushed and twisted c-channel frames in my day

The last one is just a guideline: When building something to survive impacts it must either be strong enough to withstand it or flexible enough to not care.

[pfreivald]

Solder is better than quick disconnects.
Electronics should be well-protected from 'vigorous interaction'.
Don't skimp on material where you need it (this is largely an experience thing).
Drive it like you stole it, before you bag it!

Make it out of steel.

/thread

[MrForbes]

Originally Posted by RoeeVulcan

I opened this thread in an attempt to learn from other teams experience, guessing they had a lot to teach me. I would love to know what your team did this year to a assure themselves on a robot that would come out of the field the same way it got in

i understand what you want to do....you want to learn from the experiences of other teams. I think that is great, and this forum is the best place to do that.

But I also know that the best teacher is experience. I know a lot, most of it is learned from making my own mistakes, some of it from watching the mistakes that others have made. I still have so much more to learn, and I know that the only way I can learn it is to try, and keep learning as I make more mistakes, and have a few successes along the way.

At Championships this year we had a sort of major structural failure of our robot. I had anticipated this failure, and had thought of making a brace to prevent it, but never got around to it because the robot was working OK and we had plenty of other things to worry about. The competition here was really tough, and our drive team had to drive the robot really hard just to try to keep up. I also discussed this with several of our students, and they seem to have a good understanding of how important it is to try things, and see what works and what doesn't, and make sure to learn from the failures. Failure is the best teacher.

The trick is to figure out how to learn from from others' failures, or better yet, how to experience failures of your own in a way that do not really hurt, but teach you well. Many of the top teams spend many hours trying different ways to do things, and they test over and over until something breaks or jams or doesn't work right, then they change the design and test some more. It's called 'iterative design'. Having the will to keep working on something over and over is what seems to separate the best from the rest.

Another thing you can do is to post pictures of your design here on Chief Delphi, and ask for constructive criticism. I know that there are many very experienced people here who can probably spot potential problems quickly, if you provide enough detail for them to see what you are planning to build.

[MARS_James]

Originally Posted by RoeeVulcan

So I was wondering, what do you pro teams do to build stronger robots, that can stand the thrill of the fight?

Some simple things that we have always done:
If you have a spare functioning robot with bumpers put bumpers on your robot before bag day and just have it sit there and get slammed a few times by your old bot. This will simulate rough collisions, and match play and also allow to see if your robot is top heavy and prone to tipping. If you lack a spare robot another option would be slamming it into something sturdy solid and unimportant.


Anything that exits the frame perimeter must meet one of these two criteria:
Is solid enough to withstand a robot hitting it, and can be quickly fixed if bent or broke.
Be flimsy enough to move out of the way of another robot.

For electronics your best bet is slow and steady. Once you know what you are building and how it will look give your electronics team the dimensions for their board and let them start building it so as you get done with your robot sections they can begin laying out wiring exiting the board. This serves two purposes, it gives the students who like electronics something to do during the whole six weeks and not just the last two, and by giving them time they won't rush which will allow for connections to be solid avoiding com issues, watchdogs and other not so fun stuff

My final suggestion is make sure your drive team knows what the robot can and can not survive. Our current robot is small so we could use thick aluminum this year for our structure and everything above it was made from thick sturdy PVC board that has now gone through two regionals and championship without a bend or crack with heavy defense being played on us when ever we go out. However last year we were long and tall so when ever (rarely) someone would play defense on us we had to avoid and evade so to avoid unnecessary damage.

[Golto]

Consider your center of gravity. A lot of damage comes from tipping (or falling). The bumper rules are in place to help alleviate the hit damage, worry about not damaging yourself and you'll build a fairly sturdy machine.

[DonRotolo]

Being in MAR, we go through up to 18 matches in each of three events (=48 matches) plus CMP and maybe a regional too... we need to hold together for maybe 80 matches! So building durable is required.

If you can, build a practice robot. It does not need to be perfect, just about the same as the production robot. We often use a previous year robot. Then all the prototype mechanisms get mounted.

Then beat the snot out of it. Slam it into walls, hard reverses and turns...try to break it. When it does, fix it stronger. Repeat until you are almost out of time.

Then build the final robot. Remember, it does not take as much time to fabricate the robot as it does to design it, so for the production robot we KNOW what needs to be built, it does not take long to build it.

As you can guess, it does not break often.

[Jeffy]

The most common thing to fail is the drive base. We try to complete it by week 2 or 3 and spend any free time slamming it into walls. Err... I mean programming autonomous.

[s_forbes]

Originally Posted by Jeffy

The most common thing to fail is the drive base.

That depends... We had exactly 0 issues with our drive train this season (first time that's ever happened). We used the AM kit frame.

A good way to minimize the possibility of breaking things is to minimize the number of things that can break. In general, simpler designs will need fewer parts and end up being a little lighter, so you can use that saved weight to build with beefier materials.

[cbale2000]

"How to make your robot withstand the beating"

3/8" polycarbonate... just sayin...

http://goo.gl/8UNqE

[runneals]

Our chassis survived the year pretty much unscathed. We machined ours out of tube aluminum stock (which can be found on McMaster & local aluminum retailers). This provided rigidity, along with a place to safely put 2 Clippard Air Tanks & our drive train modules.

[BrianArneson]

Build and experiment with different types of chassis and drive types (omni drive, swerve drive, tank drive, 6 wheeled etc...) and work out the problems that arise then when you got the time to fix them rather then during the build season. That way when you go to build the real one for that seasons robot you know how to assemble it. Usually it's pretty similar each year so you get good at it after a while. Summer is also a good time to use CAD programs to stress analyze where there are weaknesses so you can strengthen it up. Also, think triangles they are strong. Knowing when to use which type of nut and not leaving them laying around on the robot, it'll save you the panic later when one falls out and you don't know where it went. Cover Victors and Jaguars when working on robot, or use Talons, they can be a pain when they stop working and you have to take crucial time troubleshooting. Build simple east to fix mechanisms. This year not one thing broke on our robot because it was a simple dumper with 2 sides to hang, but still effective.