Lesson 10.1: Rapid Prototyping to Answer One Question at a Time
Technical Context
A prototype is an experiment built to answer a specific question. It only needs enough structure and accuracy to produce a useful result.
If you build the prototype like a competition mechanism, it takes longer and becomes harder to discard when the test fails.
State the Question First
Before building anything, write down the question in one sentence.
Good prototype questions are narrow and answerable:
- Can a compliant wheel at 200 RPM pull this element off the floor?
- What funnel angle acquires an element that arrives 2 in off center?
- Does a three stage cascade lift 4 lb with this spool and motor?
- How long does it take a driver to align at the scoring position?
Bad prototype questions are broad and unanswerable in one build:
- Does our intake design work?
- Is this a good mechanism?
The narrower the question, the faster the build and the clearer the result.
A prototype that changes the wheel compound, the funnel angle, and the motor speed together tells you that something improved and not which thing. Change one, measure, then change the next.
Build Only What the Test Needs
The prototype should be the cheapest thing that answers the question.
| Instead of | Use |
|---|---|
| Machined aluminum bracket | Wood, scrap plate, or cardboard |
| Precision mounting | Clamps, zip ties, tape |
| CAD and fabrication | Whatever is on the shelf |
| The real motor and gearbox | Any motor that gets close to the right speed |
| A finished chassis | A board on the floor with the mechanism clamped to it |
If the prototype needs to be accurate in one specific dimension because that is the question, make that one dimension accurate and let everything else be sloppy.
The result needs to arrive early enough to change the design. If the prototype will take more than a day or two, narrow the question or simplify the fixture.
Make It Adjustable
Since the point is to learn, build in the ability to change the thing you are studying.
Slotted holes, clamps instead of bolts, and a stack of shims let you sweep through a range of values in one session rather than rebuilding between each trial. A funnel prototype with a hinged wall and a protractor answers "what angle" in twenty minutes, while a fixed-angle prototype answers it in three sessions.
Test With the Real Element
Use the actual game element or the closest official replica available. Weight, surface texture, compressibility, and motion affect the result, so a convenient substitute can mislead the test.
Test in the real conditions too: on field tile rather than a smooth table, at the real height, with the real approach angle.
Know When to Stop
A prototype has done its job when the question is answered. Three outcomes are all successes:
It works. Record the numbers and move to designing the real version.
It does not work. Record why, and go back to the concept list from Module 0. A concept eliminated in two days is a large win.
It works but reveals a different problem. Write down the new question and plan the next prototype.
Do not keep refining a prototype after it has answered the question. Record the result, take the fixture apart if you no longer need it, and apply what you learned to the real design.
Recording It
Every prototype gets a notebook entry in the four part format from Lesson 1.2: the question, what you built, the measured result, and what you decided. Photograph it before you take it apart, with a caption.
These entries give the team a useful test record and show judges how the design changed in response to evidence.
Fill-in-the-Blank Practice
- Before building a prototype, the team should write down the single
__________it is meant to answer. - A prototype should be built from the cheapest materials that answer the question, and should be
__________so values can be swept in one session. - Prototypes must be tested with the real game element because weight, texture, and how it
__________all affect the result.
Show answers
- question
- adjustable
- tumbles (behaves, moves)
Exercise
Pick one open question about a mechanism on your robot. Write it as a single sentence, then build something in under two hours that answers it. Record the result in the four part format.
Ready to move on?
Only mark complete if you genuinely understand the material. Your progress will be saved in this browser.
Stuck on this lesson?
Ask about anything on this page. It can see which lesson you have open and which part you are reading.