How to design efficiently

Hi everyone,
I would love to get some input on how your team designs in groups efficiently. My team tends to always split into two main groups, one with veterans and one with rookies, and we design separately. Once we come together to share our ideas, we listen to the rookies’ ideas but never really implement them.

This year, we’re thinking of either splitting into 5-6 groups, each with one member of leadership and a mix of veterans and rookies, or having the veterans design and make the rookies do research. What does your team typically do? And what do you find is the most effective way of designing?

This is essentially my bible.

There are a number of resources put together by many an amazing veteran team (1114 for example) that breaks down effective design strategy but here are some of my basic pointers I use:

  1. Read the rules
  2. Understand every way to attack the game, no matter how small or seemingly insignificant
  3. Read the rules
  4. Brainstorm as a TEAM, not individuals or individual groups to get a basic idea of how you want to attack the game
  5. Read the rules
  6. As sub-groups, PROTOTYPE…PROTOTYPE…PROTOTYPE, nothing hurts worse than jumping into a design without testing the viability first and then having to start fresh 2-3-4 weeks in

There are many unique ways to accomplish the goals of a season so take advice from many and utilize your resources appropriately.

Good luck this season!!

1 Like

Speaking from a CAD perspective, as I believe CAD is the most powerful design tool that is widely accessible to FRC teams, you need to walk before you can run. Start sketching small things or components on the robot. You don’t need to CAD the entire robot, and maybe just modeling the manipulator will get you where you need.

From my experience, great 3D CAD comes from great 2D sketches. Learning how to use constraints, construction geometry, and using parametric sketching and modeling is very important to designing in CAD efficiently and effectively.

Knowing that, and considering that the robots built in FRC are relatively* simple, it’s entirely possible that an entire robot can be designed with 2D sketches (I’ve built a few robots from autocad drawings and can personally vouch for this process).

That being said, I don’t think the step from 2D to 3D is overly difficult, especially with the resources that teams have access to nowadays, specifically component libraries, free and powerful CAD software, and easily accessible reference material.

Some tips that will greatly improve efficiency within CAD:

  • Becoming familiar with the software (arguably the most important and takes the most amount of time)
    Great designers are incredible to watch when they are extremely familiar with their software (Adam Herd’s RAMP series is a great example of this)
  • Setting variables for things that are likely to change, or things that will share a value throughout the design process
    Bearing holes, game piece compression, wheel center drop, wheel diameter, robot width, length, etc
  • Using constraints (especially equal and symmetric (mirror) constraints) to limit the number of dimensions needed
    I highly recommend using dimensions only as a last resort. If you can use a constraint that makes sense in the context of the sketch, you should.

These tips, and a lot of other things that will be picked up on over time should allow for modifications and changes to be made very quickly so less time is spent designing.

This is a little demo document in Onshape I threw together to show some of these ideas. Feel free to poke around and figure out how I set the sketches up.

*Compared to, say, a jet engine or a medical device

1 Like

A lot of teams don’t realize that one of their most valuable resources is their rookies because of their ideas. Veterans can fall into the trap of being very narrow-minded because they’ve done this enough that their ideas are more or less inspired by previous robots or previous game strategies they’ve seen. Rookies come in with a fresh mind and very little previous bias so they tend to be more adept to thinking of out of the box solutions.

I like to mix veterans and rookies together early on when it comes to brainstorming because the rookies will often think of a bunch of wild and crazy ideas and the veterans can filter them based on their viability. In doing this method I’ve had a lot of veterans tell me they were told legitimately good ideas they hadn’t thought of before (alongside 30 other ideas about flying robots and shooting lasers that didn’t quite make it past the filter). It’s a good practice to get the new students involved without throwing away their talents.

1 Like

The Compass Alliance has some pretty solid resources on this.

Compass Alliance Mechanical Design Pathway

Compass Alliance Strategy Pathway

I’d also recommend going through their ‘Build a Robot’ Resources](https://www.thecompassalliance.org/build-a-robot-resources) section for some more info.

Here are some more of my go-to links for Design Strategies:
Karthik & 1114: Effective FIRST Strategies 2016 - Championship Conference
From Design Requirements to Robot Design by Aren Hill (1296)
To Compete Consistently and Effectively: Logan Farrell FRC#118
Citrus Circuits Fall Workshops 2018 - Simple Robots that Win
1678 2016 Fall Workshops - Strategic Design
1678 2016 Fall Workshops - Mechanical Design

Brandon Holley and I worked together to make these presentations, compiling information from the above presentations and our own experiences:
Strategic Decisions to Make Your Team as Successful as Possible
Game-Breaking: How to Breakdown and FRC Game

First of all, recognize that brainstorming, strategizing, and design are three separate (interrelated) processes with different dyamics and which (for most groups) work best with different meeting populations.

3946’s process the past few years has looked like this:

  • Read and digest the rules individually and in small groups as individuals find works for them.
  • Discuss the rules as a team; get everyone on the same page. Brainstorm strategies/roles a team might play.
  • The strategy group (typically 5-10 strategically minded people) prioritize the strategies in terms of value and “cost/effort”. Meanwhile, several others re-read the rules to determine if there are any rules-based issues with the strategies, and these are folded into the prioritization. The details vary from year to year, but the final stages is some sort of traceable, quantitative method.
  • As a team, select the strategy [sometimes two, one conservative, one riskier] based on the rack and stack and brainstorm architectures to achieve the strategies.
  • Form smaller groups to develop designs based on architectures. These groups are largely self-selected, but we try to have a mixture of experience and disciplines (e.g. mechanical, electrical, programming) on each
  • As a team, present designs and select the way forward.

TLDR: brainstorming and major decisions are done as a team, deep dives into the rules and design are done in smaller groups or even individually.

1 Like

Our team usually splits into 4-5 groups (we have about 25 students on the team). There is usually a member of leadership in each group. The groups all fill out a QFD (Quality Functional Deployment), which is a sheet that you input the certain aspects of the game your group feels is most important, and then input the systems that the robot could have. All the groups reconvene and present the proposed designs, and come up with the optimal design from there.

The QFD can be found <a href=http://ronweber.wix.com/flyingtoasters#!__qfd—pput>here</a> (courtesy of the flying toasters.

1 Like

Thank you for sharing! That motivates me to share some of my own teaching of Onshape (independent of the robotics context) that I haven’t sat down to do yet. Maybe in time for the season to begin.

Just pointing to my post from last year again for anyone who wants to learn more about Onshape.