
combine the best features from tested designs into one optimized instrument, then use my model to generate new data and spot its limits.
Phase 4: Build
Do Now. Phones, cars and bike helmets were all improved by mixing the best parts of earlier versions. Name one thing you own that seems to combine good ideas from different designs.
The best final design is always the single design that won the first test.
Station engineers, today you optimize. To optimize means to make a design as good as possible within the constraints. You do it in a cycle: use data to choose features, build, test, change one part, and test again.
Worked example 1: a feature table. Ava’s team compares their three wind vanes.
| Feature | Vane 1 | Vane 2 | Vane 3 | Data says |
|---|---|---|---|---|
| Tail | small card | large foam | large card | Large tails turned to face the wind fastest |
| Pivot | pin in pencil | straw on skewer | pin in pencil | Straw on skewer spun most freely |
| Weather-proof? | no | yes | no | Card got soggy in the spray test |
The combined design uses a large foam tail on a straw-and-skewer pivot. No single vane had all three best features.
Worked example 2: testing the model. The new vane is a model of the final station vane. Jamal’s team tests it with the fan from four directions. It points correctly within one compass direction every time. Then they change one part: they move the pivot closer to the tail. Now the vane swings around and points the wrong way. That data shows how one part controls the whole design.
Every model has limits. A box fan makes steady wind. Real wind gusts and swirls around buildings. So a good indoor test is not proof the vane will work by the school wall. Engineers write limits down and plan an outdoor test.
Your team optimizes ONE instrument: rain gauge, wind vane, anemometer or barometer model. Build the combined design and test it.
| Feature | Best version from the data | Evidence (number or observation) |
|---|---|---|
| Feature 1 | ||
| Feature 2 | ||
| Feature 3 |
| Test | What we changed | Result | Better, worse or same? |
|---|---|---|---|
| Combined design | — | ||
| Change 1 | |||
| Change 2 |
Write a design claim the green team could trust.
Write a claim about which features of your instrument best meet the criteria and constraints. Support it with at least two pieces of data from your tables.
Name one limitation of your indoor test. Describe the outdoor test that would check it.
Your instruments are getting stronger. Next class other teams will critique them, and you will revise.
Exit ticket: name two features you combined and the design each one came from. Why does combining them work better?
Materials: all tested prototypes from 4.16–4.17, foam trays, bamboo skewers, straws, clay, beads, spray bottle, box fan, compass rose taped on the floor.
Setup: Assign instruments so every type is optimized by at least one team; the class station will use the best of each. Tape a compass rose on the floor for vane tests.
Listen for: teams changing several parts at once; “it worked once, so it’s done”; ignoring the constraints when adding features.
Support / stretch: Give a feature table with the rows filled in. Stretch: calculate the percent drop in error from the first design to the combined design.