The Windcatcher

Hold a tube with one end facing into the wind and the other end sheltered behind an obstacle, and connect the two ends to a simple water manometer — a U of clear tubing with coloured water in it.

Expect a steady difference in level, and expect it to grow sharply as the wind rises.

Note that the windward reading is positive and the leeward one is negative relative to still air — the sheltered side is actively sucking.

Both signs are useful. Most people design for the push and forget that the pull is often the larger and steadier of the two.

Build a model room with a tower on top. First give it a single opening and measure the airflow through the room with a smoke source or a light streamer. Then open a second aperture on the opposite side.

Expect a single opening to produce almost no through-flow, and the second to start a strong steady current.

Air will not enter a sealed box. Ventilation is not a hole; it is a PAIR of holes at different pressures, and the flow is set by the weaker of the two.

On a still evening, chill the inside of your model tower — a bag of ice at the top, or simply test after the masonry has cooled overnight — and look for downward flow. Then warm one side in the sun and look again.

Expect downdraught when the shaft is cooler than outside and updraught when it is warmer.

Measure how the effect grows with tower height.

This is buoyancy, and it scales with the height of the column and the temperature difference. A tall shaft is a pump you do not have to buy, and it is why windcatchers are far taller than the openings need.

Run the intake air over a wet cloth or a shallow tray of water at the base of the shaft, and measure temperature before and after.

Expect a useful drop in dry air and almost nothing in humid air.

Then measure the humidity as well as the temperature, and note the trade being made: you are exchanging heat for moisture, and the lowest temperature you can reach is the wet-bulb temperature, not zero.

Evaporative cooling is free where air is dry and useless where it is not — which is why this technology maps almost exactly onto the world's desert belts.

Test three tower heads: one opening facing the prevailing wind, two facing opposite ways, and four open faces. Measure flow with the wind from several directions.

Expect the single opening to be best when the wind is right and useless when it is not, and the four-sided head to be never best but never hopeless.

That is a design choice, not a design error. Where a monsoon blows dependably from one quarter, build one-sided; where wind is fickle, accept a lower peak for a usable average.

Optimise for the worst case or the common case, and say which you chose.

They are ancient and geographically enormous. Wind towers appear across Iran, the Gulf, Egypt and Pakistan under many names — bâdgir, malqaf, barjeel — with the tradition running back thousands of years and the tall multi-sided Persian form well developed by the medieval period. Yazd's skyline is essentially a forest of them.

The best examples are coupled to the water system. A bâdgir drawing air across the surface of a qanat channel delivers air that has been shaded, then chilled by evaporation, then distributed — a three-stage conditioning plant with no moving parts and no fuel. Judge a vernacular technology as a system or you will underrate it, which is the same error as judging the yakhchāl's dome without its shade wall.

Air conditioning did not so much beat it as make it unnecessary. Cheap electricity and a compressor deliver cold air anywhere, in any humidity, on demand, controlled to a degree — none of which a wind tower can promise. Towers were abandoned across the Gulf within a generation. A technology that is free but conditional loses to one that is expensive and unconditional, until the price of the second changes.

It is being rebuilt, deliberately. Modern passive and mixed-mode buildings use wind towers, solar chimneys and night-purge ventilation, sometimes with a small fan to guarantee flow when physics will not. Masdar City's wind tower is the showpiece, but the same reasoning appears in ordinary schools and offices. The old design is not being copied for charm; it is being copied because the pressures it exploits are still free.

Honest limits. Flow depends on wind and on temperature difference, so it is variable and cannot be guaranteed. Evaporative pre-cooling only works in dry air. Open towers admit dust, insects, noise and rain, and need cleaning. They occupy roof space and considerable masonry. And they cannot dehumidify — which, as the air-conditioning blueprint at the end of this batch argues, turned out to be the thing that actually mattered.