How much do you design before kickoff?

FIRST General Forum
1
  1. How much design work does your team do before kickoff?
  2. How detailed are your designs (Sketches, Scaled Drawings, Completely CADed out, etc.)

I can think of plenty of advantages to working on designs in the off season, but it seems like some teams get rather attached to a design that they worked on in the off season, and are determined to use it even if there may be a better fit for the game. I know there must be teams that have overcome this obstacle. There’s constantly amazing designs being posted here on CD, and when competition rolls around theres still something new and innovative out on the field.

If you’re on a team that does do a lot of designing before kickoff how do you avoid getting attached to one. Do you just have a whole bunch of designs ready so that no matter what your particular strategy entails you have one that you can use?

Sometimes it seems almost advantageous to be a rookie team with no experience. We don’t have any previous designs to be attached too. :slight_smile:

~Allison

2

Our team has never had something truly designed before offseaon.

In a way we ‘design’ our autonomous program in the preseason when we learn how to do something that would improve our performance. For example, myself and another kid on our team are trying to learn how to program wheel encoders using interrupts. Since this is rather time consuming, we would rather start ‘designing’ this aspect of our robot ahead of time so that we will be prepared.

We do prototype a new transmission, because most likely the requirements for it will remain the same from year to year. When we prototype, it is almost like testing out a hypothesis: we test what we predicted would be a good design, and we observe the results. If we like the results, then we will keep the design. However, if we don’t like it or feel that something new or different should be added to it, we generally end up revising it during build season. So even though it’s kind of like designing…it isn’t really a true pre-kickoff design because it may end up getting revised.

I like to call it ‘testing a hypothesis’, rather than ‘pre-season design’, just because everything can be subject to change.

3

Right now, we’re designing systems that we could possibly use this year. We’re testing things out on last year’s robot, something that would be difficult for you as a rookie. It’s good to think of what potential issues you’re guaranteed to face during the season, and try to solve them before kickoff. If the drivetrain motors are running at different speeds (since you’re likely to be running one motor in reverse), how do you compensate so the robot will go straight? If an arm or other appendage is causing a motor to backdrive, how can you compensate? Do we want a shift-on-the-fly transmission, or is one speed enough? We didn’t use pneumatics last year, so we are now also learning how they work, and how to implement them. All of these problems involve both hardware and software, and thus require thought and input from numerous students.

We aren’t designing our whole robot yet, but you can bet that every robot will need a frame, a drivetrain, some sort of manipulating appendage, and sensors, so we’re starting on those.

4

We have never done any pre-season planning, but this year we’re going ahead and getting our designs completely ready for the season. Completely ready meaning that their ready to be built. Other than that, every part of our robot is based around the competition so has to wait for kick off.

5

Before this year our team hasn’t done much besides testing new ideas on old robots on the off season. However just recently we have begun building our robot; a simple lower frame a two-speed transmission assuming they will give us drill and chip motors this year. the major reason for doing this is because we are welding our robot which takes us quite a bit of time. Basically we are doing as much as possible with only a few reliable assumptions.

6

On my team, some things are close to set in stone before kickoff. This is because unless they drastically change the game on us (i.e. driving on the ceiling), we know how to make things that work. Examples:

Drivetrain. We use two Atwoods/CIMs to drive our robot. We’ve done so since 2002. Since we have almost always come in overweight (last year was an exception since we finally got a digital scale and massed the entire thing in Inventor, so we came in at 129.7), we can’t afford the weight of twin motors per side or transmissions (in particular we can’t afford the pneumatic pressure needed to operate them). So, one CIM to a wheel, and a skid as third point of contact. The gearbox, a gear/sprocket combination, is near perfect for our use and all that we will probably need to do to it is change the ratios for the new game.

Autonomous. Generally, our programmers are pushed for time as it is and trying to figure out how to use infrared sensors would not be much fun. We don’t much care for following lines because the light sensors can get shifted around, and if you lose track of the line, you’re going to wander about aimlessly, doing more damage than help. Because of this, we usually go on dead reckoning. We usually attach discs of 1/16" sheet aluminum to the wheels. With black stripes painted on the discs and a light sensor shining on the disc, we can count revolutions and correct on the way.

Frame. Since we use a 3-point drive, the shape of our robot is usually in the shape of a wedge. This allows us to get in places where other robots can’t, getting goals out of corners and forcing our way around other robots. Since we’re usually overweight, we normally use 3/4" x 1 1/2" x 1/8" thick or 1 1/2" by 1 1/2" by 1/8" thick tubular aluminum . I think this is right; I haven’t looked at it since last year.

None of this is completely set in stone, however. Last year we had a big debate over whether or not to go for twin triwheels in the front. We went with the three-points-of-contact when we decided to hang from the side. It turned out that with the three-points system, we could actually get up on the platform. I won’t go into specifics, but it involved pulling a wheelie and pushing us up with our pneumatic arm. We actually had infrared sensor spots marked and some drilled before we realized that we couldn’t afford any extraneous weight high up on the robot.

What usually helps is to create basic designs for what you think will be used - then lock them away in a drawer and pull them out after you’ve decided what you want to do and how. Figure out if any of them will help and if so, great. If not, figure out what will. It seems to do OK for us.

7

Team 469 has built a prototype robot to test new ideas on, instead of hacking up old warhorses that should rest in peace.

Many of the other ideas we have are just that, ideas. Many ideas are passed over the lunchtable in spirited discussions about the upcoming season, keeping us excited and in an engineering mood, without spending our lives in a workshop. We’re saving that for the actual build season.

I, personally, have many designs floating around the margins of my schoolwork.

8

None. :smiley:

9

We are testing a new shift mechanism with our spare 2004 trannys on our 2003 frame. We want to further cut the weight of the system. We went from pressure to vacuum during last build season, what could we have thought of in the off season??? This will give us 2 two speed bots to teach and practive robot operation with before we have an 2005 rolling chassis, build week 2-3 time frame.

No specific designs though because we keep expecting FIRST to turn the game upside down on us. We will look at priciples that could be applied to many area such as transmission design and research frame materials.

Pete

10

Originally Posted by Allison K
1) How much design work does your team do before kickoff?

Yup… Ditto. With the exception of reversing engineering things on the previous years robot(s) in case we want to use that same component again.
But, that’s not really design, now is it?

11

We are making a bunch of new stuff. Winch, tranny, vacum, oh and our new top secret cart, ill just say our bot will fly.

12

we “prototype” we are comparing the strength of a chassis using some 80/20 extruded aluminum to a solid drive plate which we have always used in the past. the results may influence or come in to play with the design of the new robot. This also allows us to train our newbies. Design in the off season, no, research and development, yes.

13

For starters…
1.) We’ve done a lot of designing and prototyping this offseason.
2.) Our team’s, and my own personal designs are done in various CAD programs (AutoCAD and Inventor, mainly)

You’re right that there’s a possibility of getting attached to a particular design before the build, in fact, we’re pretty attached to a particular system. It’s not necessarily a bad thing to get really good at designing/building the same type of mechanism over and over. It can save design and fabrication time at the beginning of the build, which as we all know is key to getting the robot built in time to practice. If it becomes advantageous to use a different design in lieu of the one we like, we will, but we don’t think that’ll be the case.

Speaking more for myself as a designer, you can never have enough ideas when it comes to your FIRST robot. You have to keep looking around for goodies to add to your bag of tricks, this includes looking at opponent’s robots, your own old robots, and machines in industry. Always try to think of ways to lighten something, or make it more compact, or easier to machine…

I respectfully disagree with your thought that rookies might have an advantage over a veteran team, having no designs they’re previously attached to. You’re at a huge disadvantage when compared to teams that have been refining designs since the last build period ended, especially those who have built and tested their ideas. You’re also at a huge disadvantage compared to teams who have experience successfully building mechanisms. I mean, even if FIRST changed every single motor in the Kit of Parts, it would only take an hour (at the most) for many teams who have designed in the offseason to adapt their designs to the new motors.

Working in the offseason is key to helping give your drivers a better chance at driving well in competition (because you’ll finish your robot earlier), which in turn gives you a better chance at winning an event (although we all know there’s a ton of luck involved in winning a Regional).

More thoughts some other time…

-Bill

14

I hope you are not really building your competition robot for the 2005 season. That would be a violation of the build rules. You can’t use anything actually constructed during the off season on your 2005 bot. All construction for 2005 must take place after the kickoff.

Here is the Robot build section from last year’s rules. Just plug in 2005 in place of 2004 and follow along.

5.2.5 Design and Build Rules
<R09> Teams must fabricate and/or assemble all custom parts and assembled mechanisms on the robot by the
2004 team after the start of the Kick-off. Mechanisms from previous year’s robots may not be used,
however, individual off-the-shelf components from previous year’s robots may be re-used to save the cost
of re-purchase of these parts IF they meet ALL of the 2004 Additional Parts and Materials Rules.
• If you do use previous years’ components, their costs must be included in the 2004 cost accounting,
and applied to the overall cost limit.
• You may not use previous years’ motors in addition to those provided in the 2004 Kit. You may
use previous years’ motors as direct replacements for those provided, however they must be the
same identical motor as in the 2004 Kit. [Note that 2002 and 2003 Fisher-Price motors are not the
same as those in the 2004 kit, and, therefore, CANNOT be used.]
• You may not use previous years Robot Controllers or Victor 883 Speed Controllers.

15

heres a lil-time line,

  1. kickoff comes and goes (yr 2003) - first thing noticed - objects on playing surface

  2. look at previous yr robot (2002) - wheels where a bad idea - for 2003 belts will be used

  3. design an easy access no hesitation for part removal robot - used napkings / paper plates / tooth picks / etc…

  4. test robot on mock up surface and ship

  5. compete with robot in regionals and nationals and design robot for 2004

  6. take 2003 robot drive train and make 2004 drive train with some slight alterations

  7. kickoff 2004 - first thing noticed - objects on playing field

  8. look at previous yr robot 2003 - belts very good idea

  9. apply 2004 drive train to robot and test and ship

  10. repeat all steps

as the season goes by you should already be thinking about next years robot or at least the drive train which is a major mild stone that each team faces. you can always build a structure for the robot cause that usually don’t change all that much but after your done with the frame you gotta remember to build the drive train. once those 2 are finished then you have plenty of time (6 weeks) to decide on how you want to be able to score. the one thing you don’t want to get stuck with is building a robot you can’t work in. always build something thats easily accessible so that you can remove parts with ease. for instance i’m working on a robot from 2002 - back in that time all the team cared about was getting it built now come 2 years later the motors goes and you find out that bosch don’t make the motors from 2002. your only option is to use the current motors which means these motors will take up more space. in 2002 it took me an hour - 2 hrs to do a motor change, now in 2004 since I have to use different size motors it has taken me lil over a month and still counting to put the motors in because:

a) back then no one thought the motors in that robot would go as quick as they did

b) the work area was not designed to handle a bigger motor so now the frame needs to be retrofitted.

c) the frame had little space to begin with to work in and now it has gotten smaller

since this robot needs to be retrofitted the followin needs to be redesigned

a) motor mounts ( since originally they just sat in the drill case and becuase of the size the bigger ones don’t fit.)
b) robot controller location ( only because where the old motors where it fit fine and now with the new motors it don’t fit at all ) - keep in mind no space
c) re wiring of power and pwms to the robot controller.

moral of the story - build the robot frame to the maximum length and width including the wheels / belts and the drive train. and make sure theres enough space in case you may need to change the design around. always look at a previous years robot and take its qualities into consideration. if you have a spare robot controller mount that to the frame to get an idea where you want it that is easily accessible

remember the design and partial build phase begins when the season starts and never ends

16

I’m sorry i cannot agree with this. Perhaps if you are re-using the frame from year to year, this might be a good idea, but it may also be a good idea to make a new frame once in a while. We made our robot 1 inch less than max wide, and we had major troubles manuerving in the competition. (getting up the stairs around the goal to hang was quite the problem.) thankfully we had an awesome driver, but now he’s graduated, and we’re definately going to take this into great consideration this coming season.

~Stephanie

<edit> woohoo! mai post was the first one on the second page! </edit>

17

Waht does work before the kickoff mean?!
:rolleyes:
For the last couple of years , we have failed to do much before the competition, bar fundraising. This year we have been assigned projects such as investigating the effects of center of mass, and building line follower equipment ( i have a strong feeling it will turn up again…). The aim is to have less work to do in the final week of the build, which has been hectic in previous years… (last year, we had about 5 people working constantly, doing shifts and sleeping in the workshop…)

18

What we do in the off season is test out ideas that could possibly be usefull in the future such as omniwheels, new drive train, autonomous mode etc. we just see if we can get a working prototype then if it would be usefull in the competition we use it.

19

the only thing we do before kickoff is meet to see if we think that the drive train needs to be modified. I try to tell the kids “why re-invent the wheel”. Yes we can try another drive train but what if it fails in compitition? :yikes:

20

Our team hasn’t really designed that much before the kickoff in the past, but last year we ended up building our robot after about 3 weeks after the kickoff and we had some problems. So this year we’re not gonna let that happen again and we’ve started designing different drivetrains and gearboxes. We’ll usually just sketch them out although today one of the mentors used salt and pepper shakers to create a model.