JVN Build Tip: Prototyping

[JVN]

So let's talk about that really successful robot. You know the one I'm talking about -- that one at the regional that seems like it can do know wrong. It just glides around the field almost twice as fast as our robots, picking up tubes without stopping and placing them on the rack almost effortlessly... what the heck! How did those guys do that?

The secret is simple... prototyping & continuous improvement.

On the most successful robots each mechanism is prototyped, thoroughly tested, and improved before it goes into the final design. Try it.

After you figure out WHAT your robot is going to do, and you come up with some ideas on HOW the robot is going to do it, you need to prototype your ideas.

When you're prototyping you need to focus on learning as much as you can about the concept to see how well it works. See how your design works in the real world, test how it interacts with the playing field. Figure out what actuators are needed. Test gear ratios out to see if your calculations are right. If you have two ideas and you can't decide... try them both and see which one works better.

Try it, break it, fix it, repeat...

You don't need to prototype everything, just the stuff you want to work.

Originally posted here:
http://jvengineering.blogspot.com/20...ototyping.html

[Justin Montois]

Everytime JVN posts there is great advice to be had.

One thing I want to add is about resources. It would be great if all teams had the resources to prototype three different arms and four different graders and three Minibots. But very few teams have that ability. Like John has posted on the past always take into account the capabilities of your team. If your top priority is the minibot then give that more of a priority over an arm ETC when it comes to using resources. Resources include time, money, materials, and knowledge among other things.

Good Luck!

[winglerw28]

Originally Posted by 340x4xLife

Everytime JVN posts there is great advice to be had.

One thing I want to add is about resources. It would be great if all teams had the resources to prototype three different arms and four different graders and three Minibots. But very few teams have that ability. Like John has posted on the past always take into account the capabilities of your team. If your top priority is the minibot then give that more of a priority over an arm ETC when it comes to using resources. Resources include time, money, materials, and knowledge among other things.

Good Luck!

If you're in our position you barely have enough to test one minibot! lol

I do agree though, every time you don't test a theory or concept you can't be 100% sure it won't fall down lol

[MattC9]

This is gooooodddd stuff we followed this pattern religously last year with our lift and IT WORKED SOOOOO WELL!!!!! but we didn't with our kicker and, yeah.

[Andrew Schreiber]

Originally Posted by winglerw28

If you're in our position you barely have enough to test one minibot! lol

Then you make that one minibot work the best you can make it and focus on it. There are a lot of great robots that only did one thing but did it exceptionally well. Look at 148 in 2008 as a great example. It is better to do 10% of the things 100% of the time than 100% of the things 10% of the time.

If you have the resources to only do part of the challenge then pick the part you want to focus on (because there is a need or you just think it is a good part) and do it.

[Starke]

Originally Posted by JVN

The secret is simple... prototyping & continuous improvement.

Try it, break it, fix it, repeat...

Great engineering minds think alike! Dylan and I just talked about continuous improvement on our blog!

Good is never good enough

One of the attributes that separates good teams from great teams is the idea or concept of continuous improvement. Most of the upper echelon teams do not take their first idea, their first prototype, and simply put it on the robot. They use that first idea as a starting point. Once the ideas have been whittled away to one final concept they take that and refine it. Most engineers will tell you that the first design is never the best, it takes several iterations to achieve that final product. And in our world of FIRST robotics we need to adopt this concept and work with it throughout not only the build season but the competition season as well. As the game evolves and develops so must our robots in order to keep up with those elite teams and find a way to bring the entire level of competition up with us. Almost every mechanism of your robot can be advanced throughout the year. By testing and practicing you can see what works and what doesn't and how you can improve upon what does work. The more testing you do the better off your whole robot will be not only from a field team standpoint but from a standpoint of being able to change things prior to competition to cut down on your cycle time and increase your efficiency(we will talk about those two concepts later on). So please remember that although you think you may have gotten it right on the very first try and you think your product is really good, good is never good enough, there is always room for some improvement.

-Dylan

[Ian Curtis]

Originally Posted by JVN

So let's talk about that really successful robot. You know the one I'm talking about -- that one at the regional that seems like it can do know wrong. It just glides around the field almost twice as fast as our robots, picking up tubes without stopping and placing them on the rack almost effortlessly... what the heck! How did those guys do that?

The secret is simple... prototyping & continuous improvement.

On the most successful robots each mechanism is prototyped, thoroughly tested, and improved before it goes into the final design. Try it.

After you figure out WHAT your robot is going to do, and you come up with some ideas on HOW the robot is going to do it, you need to prototype your ideas.

John,

From the pictures you post, it seems like 148 moves directly from cardboard and Vex into sheet metal and the production robot. Is this the case? For the robots I've worked on, we typically went from cardboard to plywood to plywood & shafts powered by drills to rough cut aluminum then to the final robot part. It seems like you skip by that rough aluminum step. I know you use a lot more CAD than we did, but do you just trust that your Vex and cardboard dimensions are "good enough"? Or do you end up rebuilding some manipulators if you find that your dimensions don't work as well as you anticipated?

[Aren_Hill]

Originally Posted by iCurtis

John,

From the pictures you post, it seems like 148 moves directly from cardboard and Vex into sheet metal and the production robot. Is this the case? For the robots I've worked on, we typically went from cardboard to plywood to plywood & shafts powered by drills to rough cut aluminum then to the final robot part. It seems like you skip by that rough aluminum step. I know you use a lot more CAD than we did, but do you just trust that your Vex and cardboard dimensions are "good enough"? Or do you end up rebuilding some manipulators if you find that your dimensions don't work as well as you anticipated?

I know they plan in adjustable features by adding extra sets of holes and such, we may have copied this attribute when we stole their entire 2007 arm effectively....

[Creator Mat]

Originally Posted by Aren_Hill

I know they plan in adjustable features by adding extra sets of holes and such, we may have copied this attribute when we stole their entire 2007 arm effectively....

Good point. Having flexibility with mounting points on a final piece can help out later down the road if something changes. Plus extra holes = less weight

[JVN]

Originally Posted by iCurtis

John,

From the pictures you post, it seems like 148 moves directly from cardboard and Vex into sheet metal and the production robot. Is this the case? For the robots I've worked on, we typically went from cardboard to plywood to plywood & shafts powered by drills to rough cut aluminum then to the final robot part. It seems like you skip by that rough aluminum step. I know you use a lot more CAD than we did, but do you just trust that your Vex and cardboard dimensions are "good enough"? Or do you end up rebuilding some manipulators if you find that your dimensions don't work as well as you anticipated?

We prototype until we get the critical dimensions. The prototypes that merit it, get more detailed work. Last year we did a 1:1 fully functional version of the entire intake + kicker out of old robot parts and sheet metal. You can see this in our 2010 video.

Some stuff goes right from cardboard into CAD into sheetmetal...

-John