Electricity-free air conditioning | Hacker News Reader

Electricity-free air conditioning

economist.com

76 pointslcuff36 comments

Existing passive solar air conditioning is underutilized.

Some ideas are very old, for example: http://en.wikipedia.org/wiki/Solar_chimney

I can see how something like that would work in arid conditions, but in areas where it is humid, it seems like you would get a lot of problems with mold and dampness.

None of the A/C alternatives I've seen on HN in the last couple of years even try to tackle the problem of humidity. They only focus on lowering the temperature, and sometimes even increase the humidity while doing so.

Maybe this has something to do with the fact that a lot of U.S. tech startups are located in relatively arid parts of the country, such as the Bay Area. I'd like to see what they can come up with if they were all relocated to Florida, or even better, Southern China.

I think i have read about older buildings that possibly combined this with stairwells.

The Tower at PNC Plaza, a 32-floor building under construction in downtown Pittsburgh, uses a solar chimney and other passive air handling methods to significantly reduce energy consumption:

http://www.paladinoandco.com/case-studies/tower-at-pnc-plaza...

Not sure why everyone is trying to couple this to solar panels. The TL;DR for the article is a material that is 97% reflective to incoming light and also a heckuva radiative source. As such, it bounces away all the sunlight and radiates additional energy out into the void. The international space station uses a similar setup to achieve this- there's nowhere for waste heat in the space station's solar panels to go (there's no air to accept it) so radiation is the only method of dumping it. Nobody does this on Earth, because radiators are heavy and take up a lot of space. It'd be far cheaper to just install more solar panels.

Solar panels try to absorb all the light they can; that 10-20% is sunlight to electricity, and most of the rest is absorbed as heat.

If you tried to combine the two by stacking, a reflector/radiator with an absorber, you'd render one or both useless. The radiator would block the solar panel's light, or the solar panel would block and reflect the radiated light from the radiator. You could point the radiator down, which would cool the solar panel a bit (making it more efficient), but you'd be heating up whatever was beneath the solar panel. That'd be great in an open field but awful on a rooftop.

This invention's biggest application is in the existing situations where you want to dump heat. Air conditioners have large radiator sections (not sure why they call them radiators because they do not radiate so much as conduct and convect the heat away) which would be able to dump a lot more heat with a properly oriented coating using this novel material (also potentially eliminating blowers and fans).

I'd really like to see a graph of heat flux versus ambient temperature versus area. If it took 1 square meter versus 100 square meters to dump, say, 1000 watts, that would change this from a curiosity to a breakthrough.

Very cool. I love the way computers are smashing technology forward.

But the article didn't answer the big question, why this over solar panels.

Grab the energy and turn it into electricity seems to make more sense, cool the building while creating $.

I'm not sure this is the answer, but it seems likely that you can more efficiently reflect energy back into the atmosphere than you can convert it via a solar panel and then use that energy through air conditioning to cool down the space.

Instead of having to go through the whole process of conversion from sunlight to electricity via a solar panel to cooling via an air conditioner, you can just reflect most of the energy that's heating up the building in the first place.

One thing I didn't think of is, if this was really cheap to buy AND install it'd work. Solar panels and inverters are currently not economical else they be 100%ish on all roofs. If this cost little, obviously it'd be a break through.

But reading the nature article which tries to get into this I'm not convinced it's that cheap or will work better than solar panels.

Solar panels are only 10-20% efficient. The rest of the sunlight ends up as heat which presumably might end up in the building. It would be difficult to cover an entire sunward side of a building with solar panels anyway which is what you would want to do with this technology should it ever be made practical.

My question would be: why this over insulation?

I think with a material like this you would use both. Based on the article, it sounds like this material has a fixed cooling effect for a given temperature and incident radiation flux. So if you want to maintain 75F in a building and outside it's 110F -- this material would reduce the cooling effort to the same as a normally insulated building where it's only 100F outside. A substantial energy savings. (I made up the 10F cooling effect, pick your favorite constant)

You wouldn't just cover the sunward side of a building with the new material. You would cover every surface that has an unobstructed view of the sky, from any angle.

Furthermore, the surface could be corrugated instead of flat, to increase the radiating surface area. The cooling would work even when radiating at a reflection of the sky.

This would likely be used in addition to insulation. Your radiators would be thermally connected to the building through a managed, thermally conductive channel, so that the cooling effect would not continue throughout the winter, for instance.

The outside surface would also be fairly strongly coupled to the ambient environment due to convection. I suspect that coupling the outside surface past the insulation would be a loss in any case where it was hot enough outside to make people care about air conditioning.

Couldn't you just put the whole thing behind a layer that does not absorb IR or visible light, but is also a poor thermal conductor?

The article mentioned that they did that using an enclosed box to test the material. It wouldn't be much additional effort to add a heat pipe that transfers heat from the object to be cooled, through the box, to the radiator-reflector. That way, you are cooling only what you want to cool rather than the whole environment surrounding the device.

I was thinking about that too. You would need one or more sheets of something transparent to infrared 8-13 microns. Such materials apparently do exist:

http://www.infraredtraininginstitute.com/infrared-transparen...

Such a material would also have to not absorb too much sunlight.

With insulation, you are insulating your ability radiate away heat as well.

I am somewhat skeptical about this.

It sounds like they tried to apply thin-film optics for this material, given their talk of layers that are both extremely thin and precisely defined. Thin-film optics relies on interference effects between transmitted and reflected light, and enables very specific optical behavior. For example, using thin films it is possible to make a stack of layers that reflects a very specific, narrow band of wavelenghts, while transmitting everything else. In real-life, rewritable optical media such as CD-RW and DVD-RW make use of thin-film optics.

The point though is that thin-film optics only works for coherent light, such as laser light. Sunlight is not coherent, and you will not get the desired interference effects, because the light waves are not "synchronized", as it were.

So while I could be wrong, I wouldn't be surprised if the performance of this reflective sheet is very low, and nowhere near what would be required to replace air-con.

I don't think the article ever implies that this would REPLACE air conditioning.

Last sentence: "But the idea of even part of a building’s cooling system being electricity-free is an attractive one, so this may be the start of something really cool."

Layers of films can be stacked to produce a broader absorption band. Since they aren't concerned with transmission then essentially they can build as many layers as they want to block the spectrum of interest (ignoring cost, of course). Perhaps a larger concern is the dependence on thin-film interference to the incident angle of the light.

> enabled the siesta-free working habits of the temperate regions

Frankly, that is an absolute downside.

I live in a very temperate region and having switched to freelancing i find that the daily siesta makes me much more productive.

Enabling is not enforcing. No downside.

In practice it is enforced in my country to forego siesta, as long as you are not self-employed.

Outsourced freelancers charging ridiculously little has also no downsides for freelancers from developed countries, globalization has just enabled a bigger workforce to participate remotely.

It would probably make more sense to use it to concentrate solar energy on to solar panels. That plus a combination of better insulation, triple glazing, and increased thermal mass would make better use of the energy. Many of the buildings that have window air-conditioners like the one used as an illustration in the Economist article are so badly designed from a thermal perspective that they would not benefit much from such mirrors anyway. A lot of the heat they suffer from is generated internally. Another way of dealing with the problem is to use the heat pumps to dump the heat into water to use for washing and process heat instead of just throwing it away. As far as I can see the mirror is exactly the same idea as a window air-conditioner except more efficient, that is, it simply moves the problem elsewhere. Still, as a component in a system of measures to control temperature it sounds like a good idea.

Solar panels are very expensive and can not be used on windows.

"Solar panels are very expensive"

Times are a changing, $1/watt stereotypical 100 watt panel is about 2x4 feet.

A window of decent quality , somewhat energy efficient, of 2x4 foot size is about $300.

TV's sell for about twice what a comparable surface area window costs. A 50 inch TV has about the same surface area as a 100 watt solar panel.

Aquariums cost about the same as windows (2ft by 4 ft front would probably be about a 60 gallon, which is about $300)

Another way to put it, is if you're looking at glass, per square inch, the cheapest (aside from picture frames) is solar panels, then windows and fish tanks cost about 3 times as much as solar panels, then TVs cost about six times as much as solar panels, per sq inch.

Back on topic you'd be far better off installing perhaps 3x to 10x the square footage of solar panels than installing very expensive windows with this film.

At least theoretically you could apply the film to siding, maybe even roofs, and have slightly cooler surfaces.

> 2x4 foot size is about $300

2 micron thick coating probably costs $0.1 for the same size.

My dad did a simpler version of this a few years ago: he used to live in a trailer with a black roof; one summer he painted the roof silver and the inside got noticeably cooler.

That's half of what is going on with this material. A silver/shiny roof will reduce the amount of solar energy absorbed into the home. But it will also reduce the heat which is radiated out of the house. So this method won't cool the house it will just insulate it better from the sun. (Which will result in cooler temperatures, though)

Something (e.g. a color, surface finish) which absorbs heat efficiently also gives off heat efficiently and vice versa. That's why steel/iron motorcycle cylinders are often painted black. The amount of heat absorbed by the sun is much less than the amount of heat which can be radiated away.

What these researches are trying to do is optimize this reflection/absorption ratio for the various light/infra-red frequencies. Pretty cool idea, I think.

This reminds me of the technology that generates electricity by emitting photons at night. You might think it were a joke if it weren't from Harvard:

http://www.seas.harvard.edu/news/2014/03/infrared-new-renewa...

So this material reflects light at a specific frequency. Could this type of reflection be used to make solar panels more efficient?

I can see this material being useful on roof terraces but in high rise buildings less so.

However I still don't understand why people would use this over a solar panel setup (cost I would assume).

I wonder how it compares to this:

http://www.jetsongreen.com/2009/06/desert-modern-rimrock-ran...

They actually had to get special approval to downsize the AC units.

I don't get it. How is this different from going under a shade?

A shade heats up and radiates long wave radiation from the back side. This would keep the shade material cool.

what if there is water condensating on the panel ? what if the panel is not perfectly oriented towards the sun ? (dunno, just asking)

Don't know what happens if there's condensation. As for orientation, here's the paper.

http://www.nature.com/nature/journal/v515/n7528/full/nature1...

You're looking for Extended Data Figure 1b). It shows relatively constant emissivitiy from 5 through 60 degrees of incident light.

Speculating on water condensation. This would probably lower its efficiency. That said, they only claim cooling of ~40Wm^-2, corresponding to ~5 degrees below ambient. Realistically, this stuff is most useful when ambient is above ~20-25 degrees (this is when most people want cooling), where the drop in temperature by ~5 degrees shouldn't cause too much of a condensation problem. Of course, that's just one application.

Reply on news.ycombinator.com