I recently had a discussion about ACs and how they heat up cities.
Then I found an article about theoretical increase of efficiency of acs by using the heat pulled from a room to run a thermoelectric device and getting some of the energy back that was used in the ac.
I‘ve had this downstream thought many times already: since hot air is basically just energy stored. Could we theoretically pull (all?) the energy from the air (depending on desired temp) to cool it and casually fuel our society’s energy needs?
You can’t output more energy than the one available in a system.
The real solution to cool off cities is trees and less pavement :)
I agree. It’s thermodynamics, right?
The reason I’m asking is that 100 m³ of 60 C air would have a specific amount of energy (watts?) in them, right? And from there to absolute zero (0K) would be “available energy” in my perception. Or is “available” something else?
Thanks for elaborating. :)
a specific amount of energy (watts?)
Energy is measured in joules. Watts are joules per second, and a measure of how quickly the energy is being used.
And from there to absolute zero (0K) would be “available energy” in my perception.
No, it’s not available. The only way to use heat energy is to find something that’s colder, to be able to transfer that heat to, and use that heat transfer to drive some other process that puts the energy in another form: in a chemical bond, in an electrical charge, in a moving object, into moving something heavy higher, etc.
Once everything in the universe completely evens out in heat, where none of the heat can go into anywhere else (because everything else is just as hot), that’s known as the heat death of the universe.
So if you’re starting with stuff that’s all the same temperature, and you want to make one part of that system colder by pumping heat out from the place to be cooled and dumping that heat into an already hot place, it’ll always cost more energy than you can capture again when you try to use that heat for other stuff. That’s because if you want to use that heat energy, the only way to do it would be to take advantage of the heat differential between the hot zone and the cold zone, by equalizing the temperature between two zones. Well, if you’re going to do that, then why did you spend energy cooling the cold zone in the first place? It’ll cost more energy to capture the heat as it returns to the cold zone than it cost to make the cold zone in the first place, so it would’ve been more efficient to just let the two zones remain equal temperature.
You would need an enormous amount of energy to achieve 0 K (-273 °C). See the system here is the atmosphere so you can think about the average outside temperature as the “state of least energy”. So you actually need to use a lot of energy to achieve that because you are going way further (although in the other direction- negative temps). Our system is earth, so 0K ain’t easy chief - check quantum computing (we need almost 0 K to work and those are huuuuge resource intensive machines). If you were in between the emptiness of galaxies, then that would be indeed the “normal default” which entropy would “go towards “. Basically, “our” entropy has a different temperature goal, because we are the system that is fed and bound to the sun. I can’t explain better because I also have limited knowledge, just the basics, sorry. Also the “” is to explain better, do not quote those as scientific.
Energy is only “available” when there is a region of higher energy density and a region of lower energy density, that you can extract work from by allowing that energy to flow from the former to the latter until they are equalized, at which point no further energy can be extracted from that system.
In the case of air conditioning, you can make heat flow “uphill”, so to speak, by applying additional energy from outside of the inside air / outside air system, usually in the form of electricity generated at a power plant. In the very large picture, though, it’s all just moving energy around from other regions of higher and lower densities, a losing usable energy with each transfer. That’s what entropy means.
Veritasium did a really good video on this idea a couple months ago, if you’re interested: https://youtu.be/DxL2HoqLbyA?si=bru50t1VYEKXKmKX
Entropy says no.
Energy likes to be heat and the only way to get heat energy out of something is by having a temperature differential. ACs spit out air that’s hotter than the environment so you could theoretically turn some of that back into useful energy, but the cost of doing that outweighs the benefits.
What’s the cost if we pipe the hot air through a steam engine?
Heat pumps (like AC units, fridges, etc) become less efficient the greater temperature difference they have to pump the heat. So pumping heat from a 25°C room to a >100°C steam engine would become terribly inefficient. It would need more energy, which creates more environmental damage and climate crisis to source, and that energy heats the cities even more.
The only sane way to cool cities is to get rid of as much concrete and asphalt as possible (especially the vast amounts of ground that is covered for cars), and keep only narrower sealed paths for small individial transport like bikes. Plaster everything with trees and grass and other greens. They cool down the city dramatically and are able to take up the water that comes down at extreme weather events.
Escaping the urban hellscape cannot be achieved by building more stuff and throwing more energy at it. Just visit a park in your city and observe how the temperature changes, it is that simple. Mobility cannot seal all surface area, it has the be minimal, i.e. narrow paths and trains with rails that can also run on open ground / green areas. This implies of course not building secluded areas for living, shopping, working etc… It has to be a mix, where commutes are short (i.e. like european cities, not american ones).
https://en.m.wikipedia.org/wiki/Organic_Rankine_cycle
That’s basically what this engine cycles doing. It’s taking heat from some source and trying to save as much energy as it can. Looking at efficiencies around 40 to 60%.
The greater the thermal gradient the easier it is to produce useful energy from it.
Sad but thanks for letting me know. Makes sense that there is some reason for it dissipating as heat most of the time.
But there are actually studies done to find out if energy can be harvested from the temp differential in acs https://www.sciencedirect.com/science/article/pii/S2352484722019023
So I suppose there might be some use to it.
There is a lot of bogus “science” out there, and this is part of it.
You need a temperature differential to harvest electric energy. You also need a differential to get heat energy to flow (usually from inside your apartment to outside). If you have that differential, you do not need an AC, you just open the window. If you do not have a differential (or if it points the wrong way, i.e. outside is hotter than inside), you need an AC + energy to create that differential, that lets thermal energy flow from your room to outside. There’s no “free leftover differential” in this, the differential it creates is literally to transport heat energy = why you have turned on the AC. Every bit you use of this differential for harvesting energy, you could turn down the AC a notch and have it save more energy than you could possibly harvest.
This idea is as mute as mounting a wind turbine to your electric car to “harvest” the headwind from driving
I understand. Thanks for explaining.
What still eludes me is if I have 35c outside and want 25c inside, those 10c times cubic meters go outside in the form of 50 or 60c hot air. And I think thats why people like me think it is possible to get those 15-25c difference back in the form of electricity (or at least not spend them so that the excess heat is not pumped outside).
Those 15-25c difference exist to drive the heat energy from the radiator into the air. If one would want to not waste as much energy for this, the actual solution would be to use a bigger radiator that can dissipate the same heat energy per h while being lower temperature. That would need way less additional material and be way more efficient than building another harvesting machine.
It is about temperature differential. If you had a cold well then yes you could heat up the cold well and extract energy in the process. But if you had a cold well why are you not already using that as part of the AC.
By the way, that is why AC using geothermal heat pump is way more energy efficient then air. Another cold sink is the sky specially a cloudless one. Another one is a heat pump hot water heater. You pump the heat into hot water that your going to use anyway cooling the room at the same time.
So physics, yes energy/entropy stuff but then it is a question what can you do. My only point, look for the loopholes too… that is what can you do.
Heat difference is what you can get energy from, not heat itself. You need something cold to get energy from the heat.
Makes a ton of sense but how does that translate to burning coal for example? You just set it on fire and it churns out tons of energy. I suppose its stored „organized“ energy which then gets released and is allowed to increase its entropy.
If you only had access to the coal furnace you couldn’t make power. The coal furnace is hot and it’s surrounded by room temperature air. The furnace really wants to heat the air around it and the air wants to cool the furnace because nature generally doesn’t like large differentials. So what we do is we force that heat to turn an engine before it can get to the cool ambient air.
It’s like a putting a turbine in the way of a waterfall. The water wants to fall, so we force it to turn an engine before it can get to the ground.
So back to your initial question, an AC is a heat pump. It pumps heat from the cooler inside to the warmer outside. It’s just like if we pumped the the water from the bottom of the waterfall to the top. Yes you can than use that water to generate energy, but you’re the one who pumped it up there in the first place so it’s a bit counterproductive.
Thats an awesome explanation! Thank your very much!
So, from this and many other comments and some independent reading on my side, we‘re technically just walking batteries getting fed by the sun, being buried under ground after dying and becoming coal so to speak.
So, theoretically, we would need to build some way to exhaust the excess heat into space (and could also get work done in the form of electricity) if we wanted to use the current overheating earth to our advantage while cooling it off. Thinking of a giant ac at this point. :D
But jokes aside, this means that the average laypersons idea about „energy“ is false. We need „work“, not energy. Because the dissipated energy can not perform work anymore. Correct?
Sounds like you’re on the right track there. As far as energy goes, you’re right, when things are dissipated, or all the same, you can’t extract anything. You need a differential, like a hot place and a cold one, a high voltage and a low one, a fast object and a slow/stopped one, a high object and a low one. The higher the differential the more you’re going to be able to extract. If it’s too small you might not be able to get any useful work out at all.
Thanks! That’s good to hear.
The problem with this is entropy. Heat dissipated to the environment is highly entropic (more disorganized, less useful heat/energy), and to effectively extract it, you’d need to reduce the entropy, which requires more energy that you would get from the heat you recovered.
Ok. Thanks for elaborating. Always entropy getting in my way! :)
However you use heat energy to generate other energy, it will not solve the problem of heat being increased in cities. Heat is an end State energy form.
This is why you may have heard the phrase the heat death of the universe. Entropy can be thought of as all energy being converted to heat, evenly distributed in the universe. The best you can do is move the heat elsewhere.
Heat death of the Universe happens when the heat is evenly distributed and there’s no way to move it to produce other forms of energy, like electricity.
Theoretically, heat can be turned into matter per Einstein, but we haven’t figured that out yet.
That was quite an awesome read. Thank you very much! :)
You are welcome! You’re making me blush.
Nope, you would need more energy to pull it from the air than you’d produce. So it can in theory work to reduce the electricity consumption, but not to fuel our energy needs.
some AC/ heat pump models pull the heat in the air and put it into water. those have existed for quite some time. look for “air to water heat pump”.
of course, there is some residual heat from the operation of the machine itself. and once the water is hot enough it will continue to release the heat into the air, but they do release way less heat into the air than regular AC units.
Could we run a water line through it from the water heater and use it as a supplemental water heater? At least we would be using that energy for something then.
they already sell appliances that do that. Daikin Altherma line, for example, can combine A/C, sanitary hot water and home heating.
In summer it takes heat from the inside of your house an puts it in the water tank. In winter, it takes heat from the air in the atmosphere and puts it both inside your home and water tank.
So we can attach a steam turbine that’s driven by the water loop and connect that to the grid. Then we can keep pumping heat into the water loop. Right? 😂
I like the way you’re thinking. There’s more than one way to cool something. Air conditioners use the power of compression of certain materials to move heat from one side to the other. But there’s other ways to transfer heat. Just like a computer you could use water cooling. You have a reservoir of water, you can cycle that through areas that are hot to cool down the area. Using your water reservoir as a massive heat sink.
You can harness the wind to help evaporate of cooling be more efficient.
Environmentally friendly air conditioning design is a very interesting subject.
Thank you! It’s interesting indeed.
Currently, the most efficient way of cooling I know of in computers are vapor chambers, so instead of a copper block, you have a chamber (also made from copper) with 1/3 or so filled with water and the lowest point is the hottest part. The water boils, evaporates and settles in the porous walls of the chamber, dissipating the heat and flowing back to the bottom. Those then go into giant fins and are transferred to air. The Nvidia a100 does this in an interesting way. They built one giant vaporchamber system into it, transferring all the heat to the top. I presume its a lot more efficient than water but I lack the specific knowledge to judge the precise efficiency.
In any case, I assume there will be more efficient ways to transfer heat in the future. What eludes me is why we always have heat output and are unable to just transfer heat to electricity or at least movement. I think I‘m missing a key information.
Heat by itself is chaotic. It is the brownian motion of molecules vibrating. If you’re in a hot room, using that vibration to get something useful done can be difficult. Try to think of a way to get those randomly vibrating molecules to do some form of physical work.
When we talk about heat gradients we mean there’s one side that’s hot I.e. moving a lot, one side that’s cold, not moving at all. One of the laws of thermodynamics is things tend to stabilize, so the hot side and the cold side are going to mix they’re going to want to move into each other and find an equilibrium. You can use that preference for movement to do mechanical work, like turning a paddle.
This is all broad strokes. You’re trying to get organized energy, out of a chaotic system. Classically when humans use heat to generate workable energy they use the heat to make steam from water, and then they use the water to do mechanical motion.
None of these systems are 100% efficient. There’s always going to be waste heat. Energy you can’t capture. And that’s the problem. It’s the waste energy that’s making air conditioners most inefficient.
Yes, thanks. That makes a lot of sense.
@Haui there was some work done recently on improving efficiency of solar panels by putting underneath a layer that converts the heat into electricity, also cooling the panels and thus improving their efficiency. So, maybe?
Yes, right. I read that too. Pretty awesome, right? :)
@Haui but, that’s tricky, because the voltage and generated power may be quite unpredictable, and generally power network and devices do **NOT** like that. Which requires additional devices to make that power predictable, which costs resources and generates heat…
