Engineering taster kit: build a water filter (or a bridge) in 90 minutes

The clients in this kit (a rural primary school and a village council), the credits and the constraint cards are FICTIONAL: for a classroom exercise. Material costs in the attachment are planning estimates, not quotes.

What this taster is

A 90-minute hands-on engineering taster for round 2 of the Day 2 job tasters. Learners build something physical, test it, measure it, and rebuild it when the rules change. It suits learners curious about engineering, the trades, environmental work and lab science, and it is a good test for learners who think they "don't like" technical work.

The journey template uses this kit twice. In the four-labs model it is the Build and Engineer taster, built around the gravity water filter (Variant A). In the eight-areas model it is the Engineering and Trades taster, built around the load-bearing bridge (Variant B), which also works as the filter's backup when water is not practical at the venue. Both follow the shared taster run sheet and evidence contract, so facilitators can compare learning across stations.

The brief for Variant A (read aloud at minute 0)

"SK Kampung Contoh is a made-up rural primary school. After heavy rain, the water in its garden and washing tank turns brown. The parent-teacher group wants a cheap gravity filter, built from things they can buy locally, that makes the water clearer for washing tools and watering plants. Your team has 15 credits to spend. You will build, test for clarity and speed, and improve your design. One rule above all: this water is never for drinking."

Why the safety rule matters

A homemade sand and charcoal filter can make water look much clearer while germs remain. The US Centers for Disease Control and Prevention describes boiling as the most widely used and effective way to kill disease-causing germs, and recommends bringing clear water to a rolling boil for 1 minute^[1]. It also says none of the household methods it lists make water with fuel or toxic chemicals safe^[1]. Use this to open a real engineering conversation: making something work is not the same as making it safe, and engineers are responsible for knowing the difference.

Materials for Variant A

Per team of three or four (full list with planning estimates in materials-and-costs.csv):

• two staff-cut 1.5-litre bottles, sand, gravel, activated charcoal, cotton wool, coffee filters, cloth squares and rubber bands
• clear cups labelled NOT FOR DRINKING, a printed clarity card (lines in sizes 6 to 24 point), a phone timer, and muddy test water made from clean garden soil and tap water
• nitrile gloves, trays or basins under every filter, paper towels, and a mop for the room
• the team worksheet and the credit price list; constraint cards A and B, one drawn per team

Staff only: the practitioner role card and a pre-printed AI layer suggestion for the paper route, both in facilitator-and-practitioner-cards.md. Most materials cost little if bottles and cloth are collected free; check local prices before buying.

Run sheet

No AI before minute 25. Version 1 is due by minute 45 and version 2 by minute 64.

The practitioner's role

The practitioner plays the client's engineer or a water-treatment technician. They push on what learners forget: who at the school cleans or replaces the filter and how they know when, what happens after the third storm when the sand is full of mud, the cost over a year, and how the team knows the water is not safe to drink. They do not give away the best design. The role card is in facilitator-and-practitioner-cards.md.

What the facilitator looks for

Use the four shared criteria and record emerging, demonstrated with support, or demonstrated independently, with an example: two short observations per learner. These are teaching notes, never scores.

• Frames the task: plans within the credit budget and names the client's real need (clearer water for washing, not drinking).
• Uses and verifies AI: tests the AI's layer order against their own measurement instead of trusting it.
• Improves after feedback: keeps what worked when the constraint card arrives instead of starting again.
• Explains a justified next step: names a trade-off (clearer but slower, cheaper but clogs sooner) and what they would test next.

Listen for a learner who says, without being asked, that clearer water is not safe drinking water. The pre-printed AI suggestion for the paper route ends with a false safety claim on purpose, to give teams the chance to catch it.

Variant B: a load-bearing bridge

The fictional client is a village council that needs a footbridge model. Teams of three or four span a 30 cm gap between two steady tables with 40 ice-cream sticks and 1 metre of masking tape, then load it gently with closed 500 ml water bottles, one at a time, until it fails. They record bottles held, sticks used and where it failed first.

The run sheet keeps the same minute marks. Each learner sketches a shape and writes an assumption before any AI (0-10); the team plans the design and who does what (10-25); AI suggests a design and the team builds and tests version 1 to failure, checking the AI's idea against the result (25-45). At minute 45 the practitioner, a civil or structural engineer or a builder, asks what failed first and why there, how much stronger than the test load a real footbridge should be, who inspects a real bridge, and what a real site needs before work starts. Then each team draws a card: A, a third of the sticks are taken away; B, a toy boat must pass underneath, so the bridge needs 8 cm of clearance. Teams rebuild and test version 2 (55-64), explain one decision in the pod review (64-82) and reflect (82-90). At minute 82, ask the practitioner for one real example of a failure or an inspection, and how site work differs from a model.

Extra materials: steady tables 30 cm apart (or a gap marked with books), closed 500 ml bottles, and a toy boat or an 8 cm card marker for card B. The worksheet has a Variant B page.

Safety

• Staff cut the bottles; learners never handle blades.
• Clean garden soil and tap water only; never drain, river or pond water.
• Gloves on, trays under every filter, spills mopped at once, and wet sand in the bin, not the sink.
• Label every cup NOT FOR DRINKING and keep learners' drinking water in a separate area.
• For the bridge: tape any wobbly table, load gently, closed bottles only, feet clear.

Paper route

The build needs no device. Teams draw their plan and write their results on the printed worksheet, and redraw the plan after the constraint card. If AI is unavailable, the facilitator reads out the pre-printed AI suggestion and teams test it like any other idea. Learners give the two-minute explanation aloud and write the reflection on the worksheet; the facilitator photographs the sheets and both builds once devices work.

What to do next

1. Programme lead: after the venue visit (water, sinks, floor surface), confirm which variant each station runs, and buy materials a week ahead.
2. Practitioner: read your role card and bring one real example of maintenance or failure from your work.
3. Facilitators: do a ten-minute dry build together so you know roughly how fast and how clear a basic filter is, or how many bottles a basic bridge holds, before the day.