Introduction
Cooling is the fastest-growing use of energy in buildings. The International Energy Agency estimates that air-conditioners and electric fans already account for roughly a tenth of all electricity consumed worldwide, and that the global stock of air-conditioners could roughly triple by 2050 [1]. Most of that growth will happen in hot countries where grids still run largely on fossil fuels and where many households can afford neither the units nor the bills. Rising heat makes the problem self-reinforcing: every new air-conditioner exhausts heat into the street and carbon into the atmosphere.
Long before electricity, builders across the Middle East and South Asia solved the same problem with architecture. The windcatcher — bâdgir in Persian, barjeel in Gulf Arabic, malqaf in Egypt — is a tower that rises above the roof to catch the prevailing breeze, where wind is stronger and air is cleaner, and funnels it down into the rooms below, while openings on the leeward side pull stale air out [5, 6]. The historic city of Yazd in central Iran, inscribed on the UNESCO World Heritage List in 2017, is still crowned by hundreds of them [11].
Traditional towers have three well-known weaknesses. They do little when the air is still; they let in dust, insects and birds; and in very hot weather they move hot air rather than cool air unless water is added. Modern engineering literature has proposed fixes — evaporative columns, solar chimneys, mechanical assistance [7, 12] — but most are designed for commercial buildings and priced accordingly. EcoCatcher asks whether the same ideas can be delivered at the scale and budget of a single family room.
What this project delivers
- A modular design of six upgrades, each under US$45 and buildable with hand tools from salvaged or locally available materials (§6).
- An open, transparent physics model combining wind pressure, stack effect, evaporative cooling and thermal mass, running hour by hour in any browser, offline, in seven languages (§2–3).
- A global screening of 24 hot-climate cities that identifies where the design works on its own, where it needs hybrid operation, and where it should not be used (§4–5).
- A validation plan with instruments, metrics and pass criteria so the model can be tested against a physical prototype (§7).
Interactive model
Choose a city or set your own climate, switch upgrades on and off, and move through the day. Every number on this page — including the abstract — is recomputed by the same model.
Interactive model available in 7 languages; the report text is in English.
Live cross-section
Results at selected hour
Mixed-mode assumption. Simplified teaching model.
Climate & context
Modern upgrades — select to toggle
Method
The model is deliberately simple: every equation is printed here, every constant is listed in Table 1, and the complete source code is in this page. It is intended for comparing design options and climates, not for certifying a building. The day is simulated in 15-minute steps over four consecutive identical days so that the thermal-mass response settles; the fourth day is reported.
3.1 Diurnal climate
Outdoor temperature follows a cosine between the daily minimum and the peak at 15:00. Absolute humidity is held constant through the day, so relative humidity rises at night as the air cools — the physically consistent behaviour of a dry desert night. The dew point is found with the Magnus approximation [8], and wind is taken as 25 % weaker at night than in the afternoon.
3.2 Airflow
Wind-driven flow through the tower is proportional to wind speed and the square root of the pressure-coefficient difference between windward and leeward openings; stack-driven flow is proportional to the square root of the temperature rise in the exhaust shaft, which the solar chimney increases in sunlight [9, 12]. The two combine in quadrature, and the solar fans add a fixed flow in daylight.
3.3 Evaporative cooling
Air passing the wet pad approaches its wet-bulb temperature, computed with Stull's empirical formula [2], with a pad effectiveness ε. Without a pad, contact with the cool tower walls gives a small effect (ε = 0.10). The water evaporated is the sensible heat removed divided by the latent heat of vaporisation.
3.4 Room heat balance and thermal mass
Internal and solar gains G (people, devices and sun through the envelope) warm the supply air on its way through the room, giving a steady-state room temperature. The real room lags behind that target with a time constant τ that is short for a lightweight room and long for a clay or rammed-earth lining, which is how thermal mass carries night-time coolness into the afternoon.
3.5 Comfort criterion
Naturally ventilated buildings are judged with the ASHRAE 55 adaptive model [3, 4], under which people accept warmer rooms when the outdoor climate is warmer. The neutral temperature depends on the prevailing mean outdoor temperature, and the 80 % acceptability band extends 3.5 K either side.
Evaporative cooling trades heat for moisture, and the adaptive model alone would rate a cool but saturated room as comfortable. The room's vapour pressure e is therefore taken from the supply air (evaporation follows a line of constant wet-bulb temperature), and comfort is judged on Steadman's still-air apparent temperature [13], which penalises humid air and rewards dry air.
Climate zones in Figure 3 and Table 2 follow the peak-hour wet-bulb temperature, the usual screening rule for direct evaporative cooling: A up to 22.5 °C, B 22.5–25.5 °C, C above 25.5 °C.
3.6 Energy, cost and carbon
The baseline is a sealed room cooled by a split air-conditioner to 25 °C. EcoCatcher is assumed to run in mixed mode: the air-conditioner switches on only in hours when the room leaves the comfort band. The fraction of cooling load avoided is scaled to a baseline annual consumption, then multiplied by the local tariff and grid carbon intensity [10].
| Symbol | Meaning | Value | Basis |
|---|---|---|---|
| A | Tower opening area (0.6 × 0.6 m) | 0.36 m² | Design choice, single-room scale |
| Cd | Discharge coefficient | 0.60 | Typical sharp-edged opening [9] |
| √ΔCp | Root of pressure-coefficient difference | 0.55 | Mid-range of reported wind-tower values [6] |
| H | Stack height, room to tower outlet | 6 m | 3 m tower above a 3 m room |
| ΔTs | Exhaust temperature rise | 1.5 K (+9 K × sun with chimney) | Solar-chimney studies [12] |
| Qfan | Solar fan flow (3 × 140 mm fans) | 0.12 m³/s, daylight only | Free-air rating; no battery |
| ε | Pad effectiveness / wall effect | 0.70 / 0.10 | Low-cost fibre pads, conservative |
| — | Pad, mesh, louver flow factors | ×0.85, ×0.92, ×1.20 (×1.45 night) | Assumed resistance / control gain |
| G | Internal + solar gains | 25 + 30 × sun W/m² | Two occupants, devices, sun on envelope |
| τ | Room time constant (light / heavy) | 2.5 h / 10 h | Order-of-magnitude for earth walls |
| γ | Psychrometer constant (sea level) | 0.667 hPa/K | Standard psychrometry |
| eAC | Baseline AC energy | 12.5 kWh/m² per month | ≈ 1 500 kWh/yr for 20 m², 6 months |
Results
The three figures below update with the scenario selected in §2. They show what happens over a full day, which upgrades actually earn their cost, and where in the world the physics allows the design to work.
Key findings
- Humidity, not heat, decides feasibility. A 46 °C afternoon in Kuwait is easier to cool than a 33 °C afternoon in Mumbai: the dry air has a far larger wet-bulb depression to exploit, and the extra moisture stays tolerable (Figure 3). In Dubai, Karachi, Mumbai and Singapore the room still feels too hot, so the design should not be used there without dehumidification.
- The evaporative pad and the thermal-mass lining do most of the work in hot-dry climates (Figure 2). Together they cost about US$65, roughly a quarter of the full design.
- The solar chimney and fans matter most on calm days. With wind below about 1 m/s they supply most of the airflow, which is why they are kept despite small effects in windy cities such as Baku.
- Payback depends more on tariffs than on climate. Heavily subsidised electricity in parts of the Gulf and Iran stretches payback past a decade even where the cooling is excellent; the carbon and grid-peak benefits remain.
- Water is the hidden cost. In a Riyadh summer the pad evaporates about – a day — similar to a household's entire greywater output. The design therefore specifies greywater or air-conditioner condensate rather than drinking water, and the controller should stop wetting the pad whenever the room is already comfortable.
Global atlas
Full EcoCatcher design, 20 m² room.
Climate inputs are indicative design-day values for the hottest month (peak temperature, afternoon humidity, typical wind), and tariffs and grid intensities are rounded national or regional figures. They are suitable for screening, not design. For a real site, replace them with an EPW weather file and the local tariff.
Materials & cost
| Component | Material choice | Cost |
|---|---|---|
| Total |
| Component | Gulf & Middle East | South Asia | Sub-Saharan Africa | Latin America | Mediterranean · Australia · US Southwest |
|---|---|---|---|---|---|
| Tower body | Compressed earth blocks, gypsum plaster | Reclaimed fired brick, lime plaster | Stabilised earth or laterite blocks | Adobe, bahareque (wattle and daub) | Rammed earth, reclaimed brick |
| Evaporative pad | Date-palm fibre | Jute, or vetiver (khus) mats | Coconut coir, sisal | Agave or sisal fibre, coir | Aspen wood-wool cooler pads |
| Frame & slats | Reclaimed timber, palm-rib (jareed) | Bamboo | Bamboo, eucalyptus poles | Guadua bamboo | Reclaimed timber |
| Chimney absorber | Reused aluminium or tin sheet | Flattened oil tins | Corrugated roofing offcuts | Galvanised sheet offcuts | Reused steel roofing |
| Pad water | Air-conditioner condensate, ablution greywater | Bath and kitchen greywater, rainwater | Harvested rainwater | Greywater, rainwater | Rainwater, greywater |
Limitations & validation plan
An honest model states where it may be wrong. The most important limitations, roughly in order of how much they could change the conclusions:
- Uncalibrated parameters. Discharge coefficient, pressure coefficients and pad effectiveness are taken from the literature, not measured on this tower. Real towers vary widely with geometry and wind direction [6, 7].
- Steady, one-dimensional airflow. The model ignores gusts, changes in wind direction, and short-circuiting, where incoming air leaves through the leeward opening without reaching the room. Computational fluid dynamics (CFD) or wind-tunnel tests are needed to quantify it.
- Indoor humidity is approximated. The room takes the supply air's vapour pressure; moisture from occupants, cooking and plants is ignored. The humidity penalty uses Steadman's apparent temperature, which is not part of ASHRAE 55, so humid-climate comfort results remain uncertain.
- Adaptive comfort outside its range. ASHRAE 55 defines the adaptive model only for prevailing mean outdoor temperatures up to 33.5 °C; hotter cases are clamped to that limit [3].
- Idealised climate. One design day per city stands in for a whole season. Heatwaves, dust storms and humid spells are not represented.
- Water quality and hygiene. Greywater on a pad needs filtering and regular drying cycles to prevent odour and microbial growth. The design avoids aerosol sprays, but the risk is not zero.
- Embodied carbon and cost. Prices are indicative, and the carbon in the materials themselves is not yet counted.
| Stage | Set-up | Instruments | Metric | Pass criterion |
|---|---|---|---|---|
| 1 · Pad bench test | Pad in a duct with a box fan; dry and wet runs | Two SHT31 temperature–humidity sensors (±0.2 °C) | Measured effectiveness ε | Within ±0.10 of model |
| 2 · Scale airflow | 1:5 tower on a plywood room, fan as wind, incense smoke for visualisation | Vane or hot-wire anemometer | Q vs. v slope, i.e. Cd√ΔCp | Model error < 25 % |
| 3 · Full-scale pilot | One roof room, 2–4 summer weeks, tower vs. identical control room | ESP32 logger, 4 sensors, water meter, local weather station | Hourly indoor RMSE; litres per day | RMSE < 1.5 °C |
| 4 · Occupant comfort | Short survey of consenting occupants | ASHRAE 7-point thermal-sensation scale + logger | Votes between −1 and +1 | ≥ 80 % of votes |
Roadmap & future work
- Phase I · Prototype
Build and instrument a 1:5 model
Plywood room, earth-plaster lining, removable pad. Log temperature, humidity and airflow every minute with an ESP32.
- Phase II · Calibrate
Fit the model to measurements
Estimate ε and Cd√ΔCp from data, report the error before and after calibration, and publish the dataset openly with a DOI.
- Phase III · Higher fidelity
CFD and full-year simulation
Model the tower in OpenFOAM or SimScale to quantify short-circuiting. Run a full year with real EPW weather files in EnergyPlus or Ladybug Tools.
- Phase IV · Pilot
One real room, one real summer
Install on a school or community building on the Absheron peninsula and compare it with an identical control room.
- Phase V · Share
Open-source kit and publication
A multilingual build guide, a peer-reviewed student paper, and science-fair entries. Partner with NGOs working on heat resilience in informal housing.
References
- International Energy Agency (2018). The Future of Cooling: Opportunities for energy-efficient air conditioning. Paris: IEA.
- Stull, R. (2011). Wet-bulb temperature from relative humidity and air temperature. Journal of Applied Meteorology and Climatology, 50(11), 2267–2269.
- ASHRAE (2020). ANSI/ASHRAE Standard 55-2020: Thermal Environmental Conditions for Human Occupancy. Atlanta: ASHRAE.
- de Dear, R. J., & Brager, G. S. (1998). Developing an adaptive model of thermal comfort and preference. ASHRAE Transactions, 104(1), 145–167.
- Bahadori, M. N. (1985). An improved design of wind towers for natural ventilation and passive cooling. Solar Energy, 35(2), 119–129.
- Hughes, B. R., Calautit, J. K., & Ghani, S. A. (2012). The development of commercial wind towers for natural ventilation: A review. Applied Energy, 92, 606–627.
- Montazeri, H., & Azizian, R. (2008). Experimental study on natural ventilation performance of one-sided wind catcher. Building and Environment, 43(12), 2193–2202.
- Alduchov, O. A., & Eskridge, R. E. (1996). Improved Magnus form approximation of saturation vapor pressure. Journal of Applied Meteorology, 35(4), 601–609.
- CIBSE (2005). AM10: Natural Ventilation in Non-Domestic Buildings. London: Chartered Institution of Building Services Engineers.
- Ember (2024). Global Electricity Review 2024 and yearly electricity data. London: Ember.
- UNESCO World Heritage Centre (2017). Historic City of Yazd (inscription 1544).
- Bansal, N. K., Mathur, R., & Bhandari, M. S. (1994). A study of solar chimney assisted wind tower system for natural ventilation in buildings. Building and Environment, 29(4), 495–500.
- Steadman, R. G. (1994). Norms of apparent temperature in Australia. Australian Meteorological Magazine, 43, 1–16.
Cite this project
@misc{yusuf2026ecocatcher,
author = {Yusuf, Maksud and Naghizade, Jamal},
title = {EcoCatcher: A Low-Cost, Zero-Grid Windcatcher for a Warming World},
year = {2026},
month = oct,
note = {Interactive project edition, version 2.0},
howpublished = {Online}
}