Hacker News new | ask | show | jobs
by zbrozek 1753 days ago
Oh sure, there's cases where the output is more valuable than the input. I agree that desalination is possibly quite worthwhile. Though I'd imagine that in the ocean water case that the flow rate is high enough that little water is heated to the point of evaporating. Probably don't want to produce brine or salt residue.

But much of the article talks about how 40% of lake, river, and well water goes into evaporative power plant cooling. Recovering that already-salt-free water via condensation vs just not evaporating it in the first place seems silly. Your car doesn't use evaporative cooling, it uses a heat exchanger. So too could a power plant.

Though if the working fluid is steam I guess we'd call it a condenser rather than a radiator. That said, if you have steam to exhaust you have energy left to extract. My home water heater condenses the steam out of combustion exhaust to achieve good efficiency.

1 comments

It's all about the surface area. The evaporative chiller couples the heat to a much larger volume of air, much more quickly, than a water-air convective heat exchanger. This is due to the high enthalpy of evaporation of water. The hyperbolic towers get an extra boost due to convection.

Once it evaporates to steam, it mixes with more air, which cools it and the steam condenses to fog. But at this point, this heat transfer is of no consequence to the power plant. You can collect this fog without affecting the thermal efficiency of the plant.

But you can't skip that evaporation step because it's the high surface area of the sprayed droplets evaporating that does all the work. You also can't have the collector too close because then you just have a still, and the collector plate heats up from the condensation until it equilibrates. You need the air mixing.

Basically TFA describes a cheap way to get a much larger effective surface area.

So... cost, basically? A condenser or heat exchanger will be more capex but consuming fresh water is more opex. And something like these water capture panels might be a way to pick an operating point somewhere in the middle: lose a bit of water, but capture a lot of it for low up-front expenditure?