Area: Agriculture · TRL: 3 of 9 (proof of concept on paper) · Value-engineering target: USD 300 for the cooling equipment kit; kit USD 300, structure USD 574 · Difficulty: 2 of 5
Walk-in evaporative cooling chamber with a solar fan that forces air through a wetted pad, and a controller that responds to humidity.
Studio render · Open-top render · Exploded render · Interactive 3D model · Concept blueprint (PDF) · General arrangement ZBX-DWG-001 (PDF) · STEM check ZBX-STM-001 · Concept options ZBX-OPT-001 · Requirements ZBX-PRS-001 · Traceability ZBX-RTM-001 · Sizing note ZBX-CAL-001 · Prototype build plan · Design decisions · Review note
Refrigeration is the right answer for a market hub with a business behind it, but it is the wrong starting point for a farm with no grid and no capital. Evaporative cooling needs only water, air and a little power, and in hot, dry weather it can hold produce several kelvin below the outside air (about 10 K on paper for ZeerBox). ZeerBox takes the principle of the clay zeer and the brick zero energy cool chamber and scales it to a room a person can walk into, with shelves at working height, a pad and two 12 V fans for a controlled airflow, and a controller that stops the store from making things worse when the air turns humid.
It is open and garage-buildable because the people who need it are far from any supplier of cold rooms. The structure is ordinary masonry, timber and corrugated sheet that a local mason and carpenter already know, and the cooling kit is a poultry-house pad, generic DC fans, a small pump and solar home system parts, all replaceable from a regional town. Publishing the geometry, calculations and bill of materials lets a partner adapt the size, wall material and crop mix to its own site.
About 14 % of the food produced for human consumption is lost between harvest and retail, and a further 17 % is wasted. UNEP and FAO estimate that the lack of effective refrigeration alone caused the loss of 12 % of total food production in 2017, enough to feed around 1 billion people at a time when 811 million were hungry (UNEP and FAO, Sustainable Food Cold Chains, 2022).
The losses fall hardest on perishables and on the farmers who grow them. Without a cool place to hold a harvest for a few days, a grower must sell on the day of picking at whatever price is offered. A solar walk-in cold room in the Malawi and Zimbabwe pilots cost about $40,000 per unit (Efficiency for Access), far beyond a smallholder or a small farmer group.
| Industry | Use |
|---|---|
| Smallholder horticulture | Hold one to five days of tomatoes, peppers, okra and leafy greens so they can be sold when the price is right |
| Farmer groups and cooperatives | Shared store at a collection point, loaded by several members on a rota |
| Market trading | Keep stock fresh between market days at a stall or small depot |
| Agricultural extension and NGOs | Demonstration store and training unit for postharvest handling programs |
| School and community gardens | Short-term storage of garden produce for school meals or local sale |
| Country or region | Why it matters there |
|---|---|
| Mali and the wider Sahel | Long hot, dry seasons suit evaporative cooling; in field use in Mali, brick chambers cut the daily peak temperature by up to 10.4 °C when the air was below 40 % RH, but by only 4.2 °C above 70 % RH (Verploegen, Sanogo and Chagomoka, MIT D-Lab; MIT News, 2018) |
| Northern Nigeria | Home of the modern zeer revival (TIME, 2001); pay-per-crate solar walk-in cold rooms such as ColdHubs, at about $0.50 per crate per day, sit in markets, collection centers and farm clusters rather than on individual farms (ColdHubs; IFPRI) |
| India | The zero energy cool chamber was developed at the Indian Agricultural Research Institute; one study found tomato shelf life rising from 6 to 11 days (e-planet 18(2)) |
| Kenya | MIT D-Lab has piloted forced-air evaporative chambers, some solar powered, the closest prior work to ZeerBox (MIT D-Lab) |
| Malawi and Zimbabwe | Solar cold room pilots showed the demand and the cost, about $40,000 per unit, with some rooms largely unused (Efficiency for Access) |
| Australia (arid interior) | A high-income example with a long evaporative tradition: the Coolgardie safe, invented in the Western Australian goldfields in the late 1890s, kept food cool by trickling water through a hessian cover (Western Australian Museum) |
The idea traces back to the pot-in-pot cooler, or zeer, that Mohammed Bah Abba popularized in Nigeria from 1995: a small earthenware pot inside a larger one, with moist sand between them, cooling the inner pot as the water evaporates (MIT D-Lab, evaluation of evaporative cooling in Mali). Bah Abba, a teacher from a potmaking family, received a Rolex Award for Enterprise for the work; TIME reported that eggplants kept for 27 days instead of three and that 12,000 coolers had been sold (TIME, Best Inventions of 2001). The zeer shows how far evaporation can go with no power at all, and also where it stops: a pot holds a family's vegetables, not a farmer's harvest, and it cannot tell when the air is too humid to cool. ZeerBox keeps the principle and the name and asks what the same idea looks like at the scale of a walk-in room, with a fan, a deeper pad and a sensor to decide when to use them.
Smallholders lose produce after harvest because they have no cold storage. Refrigerated cold rooms cost tens of thousands of dollars, and passive evaporative coolers are small and fail in humid weather. Design with, not for: requirements must come from co-design sessions and field trials with the intended users through a local partner.
A walk-in store (2.4 x 1.5 x 2.0 m inside, 24 crates) with double brick walls around a dry rice husk cavity. Two 12 V fans pull outside air through a 150 mm wetted cellulose pad, and a controller reads temperature and humidity inside and out to choose between evaporative cooling, night ventilation and hold. A 150 W solar panel with a small battery powers it. The TRL 3 calculation puts the store at about 24.3 °C on a 35 °C, 30 % RH day, 10.7 K below the outside air, using about 43 L of water. Store humidity is about 81 %, just above the 80 % target and at risk in the unfavorable case (estimates, ZBX-CAL-001).
Full design precis: docs/02-concept.md
Every physical, chemical and biological claim was checked against known science (ZBX-STM-001, docs/04-calcs/stem_check.py): 30 of the 36 claims for this design are within known science and 6 are plausible, five of which need supplier data or field data at TRL 4 (pad effectiveness and pressure drop, shelf life, roof-space temperature, ventilation through a long sunless spell). One detail failed and was fixed under Amish's standing instruction: the load cable now has its own 5 A fuse (ZBX-DEC-001 D1). An independent reviewer tried to refute the results; its findings are in the STEM check.
Three working principles were modelled at the same detail and scored against the user requirements (ZBX-OPT-001): A, this forced-air pad store (190 of 235); B, a passive wet-wall chamber (173); C, a charcoal-wall wind cooler (173). All three are within 10 %, so the choice is Amish's, with no pick (Proposed, awaiting Amish; ZBX-DDR-004). Option A stays the design of record and is the only one that meets both the 8 K cooling target and the mode indicator.
- Double brick walls with a dry rice husk cavity, insulated ceiling and shade roof
- 150 mm cellulose evaporative pad with drip header and gutter
- Two 250 mm 12 V DC exhaust fans in a ceiling fan box at the door end
- 60 L sump drum with a 12 V pump
- 150 W PV panel, 20 A PWM charge controller and small LiFePO4 battery
- Controller with inside and outside humidity and temperature sensors
The priced bill of materials is in bom/bom.csv. Value-engineering target: USD 300 for the cooling equipment kit. Estimated cost of the constructable design: USD 300 for the kit (at the target); the structure is built from local materials and costed separately, about USD 574 (indicative).
The prototype build plan (ZBX-BLD-001) shows how to build the first store, component by component and step by step, with a making sketch for each of the 18 made components and a picture for every one of the 26 assembly steps. Making the concept buildable added a strip footing, lintels and lined openings, a framed ceiling and roof, and moved the two fans from the front wall into a ceiling fan box (decision record ZBX-DDR-003, accepted by Amish on 2026-10-02). A local mason and carpenter can build it; the cooling kit is bought and fitted. Decisions are kept in the design decisions register; the concept choice (D2) and ventilation with a flat battery (D3) await Amish.
A walk-in room: the door must open from inside. Warm recirculating water can grow Legionella, so cover, drain and clean the sump weekly. Contains a LiFePO4 battery; use a BMS, fuse the battery leads at 15 A and the load circuit at 5 A, and keep it shaded. With the fans off, air the store before going in. Guard the fans. Not for meat, fish, milk or medicines.
| Folder | Contents |
|---|---|
docs/ |
Problem, STEM check, concept options, user and product requirements, traceability, calculations, build plan, decisions, regulatory, feasibility, risk, market and plan |
cad/src/ |
build123d Python source, the source of truth for all geometry: model.py (every option at detail level D3), product_model.py (render views) |
cad/step/, cad/stl/ |
Per-part STEP files of the chosen design (cad/step/A/), an assembly STEP per option, and STL of the chosen design |
cad/drawings/ |
2D sketches and dimensioned drawings |
bom/ |
Bill of materials |
electronics/ |
KiCad schematics and PCB layouts |
firmware/ |
Microcontroller code |
media/ |
Renders, perspectives and photos |
build-log/ |
Dated prototyping notes |
Controlled documents follow the portfolio documentation standard. Each carries a document ID (ZBX-PRC-001 for the precis), a version and a revision history. Branded PDFs are built with python .kit/render.py and attached to GitHub Releases when a document is tagged, for example ZBX-PRC-001/v1.0.
Designed by Amish Chadha. See CONTRIBUTORS.md for roles. To cite this design, use CITATION.cff (GitHub shows it as "Cite this repository").
AI assistance (Claude) was used to accelerate concept renders, prototype documentation and first-pass sizing calculations. Design direction and all decisions are Amish Chadha's, recorded in this repository's decision records (docs/decisions/).
- Hardware (CAD, drawings, BOM, electronics): CERN-OHL-S v2
- Software (firmware, scripts, notebooks): MIT
A project of the Design Molecule lab.