Harnesing cold outer space on the ground through a transparent IR wavelength.

[BrianHG]

Yes, there is a narrow IR band which is transparent through our atmosphere and a passive material color tuned to that IR band, current prototype, now can cool down 5 degrees C below ambient temperature when pointing upwards to the sky without obstruction, even when the sun is directly on it.
Physicist Aaswath Raman Ted Talk Here:

[StillTrying]

For some reason I'm not very convinced.
The difference between a black and any highly polished surface could be 5 deg. with both in the sun.
Let me guess, they just need some more funding.
http://www.cleantechup.org/skycool-systems/

[BrianHG]

For some reason I'm not very convinced.
The difference between a black and any highly polished surface could be 5 deg. with both in the sun.
Let me guess, they just need some more funding.
http://www.cleantechup.org/skycool-systems/

How is it possible that they have a solid flat object 5 degrees cooler than ambient air temperature?
This was measured...  Their graph compared it to air, not a black surface compared to a mirrored surface.
It's written in the graph at 8min 50sec in the video.
If it's a flat out lie, then they will be caught very fast as any IR camera pointed at the surface + air temp thermometer should easily read a difference with today's accuracy of literally bottom end cheap junk grade measuring equipment will be within 2 degrees Celsius.

[T3sl4co1l]

IIRC, average temperature of daytime sky is around 60°F below ambient, give or take humidity.  Not that you can take advantage of that full difference under most practical conditions -- actual incoming power to be dissipated, radiation and convection from ambient...

Mind, it's about zero on a cloudy day, for obvious reasons.

In other words: for the same reason a solar oven or death ray works when pointed at the sun, it works in the other direction when pointed away.  And it works that much better at night, when the sky isn't glowing.

Tim

[JohnnyMalaria]

The presence and importance of the window have been known for a century:

https://en.wikipedia.org/wiki/Infrared_window

Yet another case of "discovering" something old and spinning it for research dosh.

[BrianHG]

The presence and importance of the window have been known for a century:

https://en.wikipedia.org/wiki/Infrared_window

Yet another case of "discovering" something old and spinning it for research dosh.

It's not that the infrared window was discovered, it about developing a modern material which makes use of it.

[Marco]

The original paper says 40W of cooling power per square meter, so about 10W of saved AC power per square meter. Or you could have 100+ W for that square meter from a solar panel.

Would have to be very cheap and durable to make much sense.

[Kleinstein]

The solar panel gives you 100 W when the sun is shining (a little more at the right angle, but less with a larger angle of incident). The sky cooling would give you the 40 W heat sink as long as the sky is clear.

One problem I see is that the 40W off cooling seem to be before having insulation losses to ambient. So if the temperature is below ambient the useful power would be lower - possibly considerable lower, so much that they only got about 10 K below ambient in there test setup. For this system to work well it need a clear sky and preferably dry air. So chances are you get more than 10 K below daytime temperature at night under such conditions.

It is a really nice physical demonstration, but I don't see a real application, especially not at day-time. It might be worth it at night, to get below the already lower air temperature. Keeping it covered with solar cells during the day would protect the plastics film from UV damage.

[StillTrying]

This latest try looks a bit suspicious to me, if the magic paint has been available for years.

"Over the hot summer months, SkyCool reduced air-conditioning operating costs by more than 50% (see graph)."

"SkyCool has been on the Australian Market since July 2003 and distribution licences for both Australia and overseas markets are currently being negotiated."
http://www.skycool.com.au/ecos_article.pdf

[Kleinstein]

The paint / coating seems to be kind of the opposite as the coating of thermal solar collectors. Just metallic (reflecting) roof and a clear coat could be a reasonable 1 st. step: most plastics / organic coatings absorb in the IR window range and the metal below reflect the rest.

That 50% reduction in air-conditioning costs really depends on the conditions. Especially just at the edge, just a few degrees lower can make you live without AC. It also depends on the construction below. That magic paint won't help much if the roof is covered with solar cells anyway.

[Eka]

Just look for very high reflectivity roof paints. Good ones are more painful to look at during a sunny day than fresh snow, but can keep a barn or shed much cooler in the middle of the day. If it doesn't reflect 85+% of the sunlight that hits it, keep looking. It should also have a 20+ year life expectancy. Note, a polished aluminum surface only reflects 50-60% of the light that hits it. A high reflectivity white can reflect 90%. A plain white will only reflect around 25%.

There is an old reflectivity standard based on highly reflective mirror surfaces. Many high reflectivity white paints now available reflect more light than the old mirror standard thought was the maximum possible.

This is a good read:
http://www.energy-seal.com/UserDyn/ACS/pdfs/ta_how_&_why_cool_roofs_work.pdf

And all rated roof coatings, sort on 3 year SRI. High numbers are better.
http://coolroofs.org/directory

[PlainName]

This has to be bullshit, surely. IR is emitted by hot stuff and whether there is something else to absorb it is irrelevant. What they're saying here - in that IR of the right frequency can get through the atmosphere where other IR can't and thus cool whatever is emitting it - is like saying your fancy LED torch will be brighter if there's no cloud cover. No, it might /look/ brighter in comparison, but clouds (or lack of them) cannot affect how much radiation that LED pumps out.

[xaxaxa]

This has to be bullshit, surely. IR is emitted by hot stuff and whether there is something else to absorb it is irrelevant. What they're saying here - in that IR of the right frequency can get through the atmosphere where other IR can't and thus cool whatever is emitting it - is like saying your fancy LED torch will be brighter if there's no cloud cover. No, it might /look/ brighter in comparison, but clouds (or lack of them) cannot affect how much radiation that LED pumps out.

This is a subtle consequence of thermodynamics; the reason why objects on earth don't spontaneously cool down due to blackbody radiation is because the surroundings emit radiation of their own, and when an object is at the same temperature as its surroundings the emitted radiation is equal to the received radiation. This is where the concept of radiation "temperature" comes from. With no cloud cover or atmosphere obstruction you "see" into outer space and thus its temperature, and an object can cool down below ambient temperature.

[Zero999]

This has to be bullshit, surely. IR is emitted by hot stuff and whether there is something else to absorb it is irrelevant. What they're saying here - in that IR of the right frequency can get through the atmosphere where other IR can't and thus cool whatever is emitting it - is like saying your fancy LED torch will be brighter if there's no cloud cover. No, it might /look/ brighter in comparison, but clouds (or lack of them) cannot affect how much radiation that LED pumps out.

It's not bullshit. It's true. It's known as radiative cooling and has been known about for millennia and was used in India to make ice, before refrigeration was widespread. It's the reason why frost can develop on a car's windscreen, even when the air temperature is above freezing, which is common in the UK, especially in Autumn and spring.
https://en.wikipedia.org/wiki/Radiative_cooling

When there are clouds infra red is reflected back to the ground, keeping it warmer, than it would be if there were no cloud cover, allowing the radiation to escape into space.

The atmosphere has different opacities and reflectivities to different wavelengths of electromagnetic radiation. The visible light from an LED torch will be radiated away into space on a clear night, but reflected back, when it's cloudy. Because the torch isn't very powerful and the light will be scatter, when it reflects back, this won't be noticeable, but it doesn't mean, it's not happening.

The best kind of surface, for cooling, would be reflective to most of the radiation emitted by the sun, but absorbent in the parts of the spectrum where solar radiation has less energy and the atmosphere is the most transparent.

It's perfectly plausible a surface can radiate heat out to space becoming cooler than the ambient, as long as the sky is clear and the sun isn't shining on it.

[CatalinaWOW]

Hero hints at the real nubbin of why this works.

The sun has its peak emissions in the visible spectrum, while objects at around 300K have their peak emissions out near 12 micrometers.    Reflection and emission have a strong tendency to be reciprocally related so being reflective to avoid absorbing solar heat tends to make you a poor radiator at those temperatures.  Similarly, being a good emitter in the 12 micron band tends to make you a good absorber there also.  But since the sun and earthly objects are separated in their peaks you do get a chance at cooling.

These same concepts are why metal objects get so hot in the sun.  They are extremely reflective at long wavelengths (can't get rid of heat), and while reflective at visible wavelengths, they are not as reflective as they are at longer wavelengths so they collect more energy than they shed until they get well above ambient temperatures.

So there are no physics denying what these guys are doing.  But the technology to make a coating with the appropriate dips and peaks in its emission/reflectivity cheap, durable and operable in the normal dirt, dust and crud of the real world is challenging to say the least.

[T3sl4co1l]

The sun is more powerful than any terrestrial object, at all wavelengths.

Actually, this can't ever work very well, if it's a passive surface thing, not reflectors.

The spectrum at the surface has notches due to atmospheric absorbance.  Which means beaming waves up into those bands, can only get you to the temperature of that atmospheric layer.

That might still be a net gain over conventional methods, but it's no radiation-to-free-space.

It's worth emphasizing this property of black body radiation, because it's often poorly shown in textbooks.  When the temperature rises, the output rises unconditionally, in all bands.  It rises the most near the peak wavelength, but all other bands (up to the cutoff) rise as well.  Indeed, the sun is extremely intense, anomalously so, at certain wavelengths: in the shortwave band, the corona has an apparent temperature of millions of degrees.  This is one way we know the energy of its particles, indeed the first way (historically), because of the simplicity of this measurement and the extreme power output (nearly millivolts (IIRC) on a not-very-high gain antenna pointed at the sun).

Tim

[StillTrying]

IR is emitted by hot stuff and whether there is something else to absorb it is irrelevant. What they're saying here - in that IR of the right frequency can get through the atmosphere where other IR can't and thus cool whatever is emitting it

I agree, their explanation of emitting through the IR window wavelengths seems like an (infra)red herring. It's like saying a bright torch will emit more light into a dark room than it would into bright daylight. I doubt if the light leaving the torch even knows whether it's heading for the dark room or the daylight.

Absorption is different, if they said that it only absorbs IR at the IR window wavelengths where there is very little IR while pointing to the sky that would be different, but I haven't seen that explanation anywhere.

Even the wiki on radiative cooling doesn't mention the IR window anywhere.

[BrianHG]

What if you develop a product which absorbs numerous IR bands and glows/convert them to that narrow IR band to radiate the heat?  Sort of like florescent colors which take in UV light and give off visible light.

Is there not also meta-materials which can shift and reflect off the wavelength of light shined on them?

[Marco]

Note, a polished aluminum surface only reflects 50-60% of the light that hits it.

A passivated polished aluminium surface, not a freshly polished one.

Absorption is different, if they said that it only absorbs IR at the IR window wavelengths where there is very little IR while pointing to the sky that would be different, but I haven't seen that explanation anywhere.

Emissivity/absorptivity generally go hand in hand.

I don't think the Australian Skycool and this Skycool are related, they just do the same thing in different ways. With the Australian one seeming the more economical.

[StillTrying]

What if you develop a product which absorbs numerous IR bands and glows/convert them to that narrow IR band to radiate the heat?

I can't see how it matters what IR band it radiates at.

[Kleinstein]

Emission and absorption are related - its the same material constant entering here.

Any body will emit radiation at any reasonable frequency / wavelength. Just the intensity is different depending on temperature and the absorption / emission coefficient, which can be different for any wavelength.
To get best cooling one would like high absorption / emission coefficient for wavelengths where there is little external radian - this is mainly in the so called IR window range. For wavelengths with a lot of external radiation low absorption is desirable.

The radiation in the IR band can depend a lot on the humidity in the air - in a dense fog it gets close to black body radiation and no more cooling. The cooling effect it best with clear sky and with a really good material might even work in direct sunlight. Here it helps that the sun only covers a rather small part of the sky. So the IR radiation of the sun is diluted enough to allow net cooling in some IR parts.

[PlainName]

I sit (almost) corrected

The effect (which I admit I didn't know is real) seems to rely on the combination of emission and absorption, but I only addressed emission since that's basically what they are describing. Maybe they should spend a few extra words to be clear about what's going on.

[Marco]

The original paper says 40W of cooling power per square meter, so about 10W of saved AC power per square meter. Or you could have 100+ W for that square meter from a solar panel.

One of the field trials of the Australian Skycool is interesting in this respect. Allegedly it saved about 20 MWh per month on AC for a 3300 m2 roof, which is about 8 Watt per m2.

If this new Skycool can only do as well as a "simple" paint formulation it has little hope of being competitive. Also if the Australian Skycool really works this well it deserves more recognition.

[cprobertson1]

One of my friends mentioned this to me the other day - it certainly does look pretty cool! (Hehe....)

According to FSEC-CR-1502-05 (Florida state contracted research, 2005) 100% humidity would result in a 50% loss of efficiency - and am I right in thinking that the efficiency is moderated in direct proportion to the cloud cover (how transparent are clouds to this particular IR wavelength?)

What do you folks think?

[CatalinaWOW]

Clouds will shut this down.