Omnidirectional air conditioner allowed by physics?

  • Thread starter Fooality
  • Start date
196
42
Summary
Can an air conditioner cool the air in all directions provided it emits energy in some other form?
I know heat is a form of energy, and that’s conserved, so usually an air conditioner will emit heat on one side, cooling on the other, probably increasing the overall heat in total.

What’s not clear to me is if it always must do this, specifically if the air conditioner produces another form of energy: For instance I’m picturing an electrical device that cools on it’s whole exterior, and uses heat inside to generate output electricity or something through some kind of heat engine on the inside. The Wikipedia page for heat engine says:

“During this process some of the thermal energy is converted into work by exploiting the properties of the working substance.”

What’s unclear to me is if the amount of thermal energy converted to electricity can be large enough that that the net cooling of the device can be greater than the net heating, or if this violates some fundamental laws. (I’m not a physics guy)

Thanks for any guidance or resources.
 

anorlunda

Mentor
Insights Author
Gold Member
6,824
3,753
Summary: Can an air conditioner cool the air in all directions provided it emits energy in some other form?

What’s unclear to me is if the amount of thermal energy converted to electricity can be large enough that that the net cooling of the device can be greater than the net heating, or if this violates some fundamental laws.
Well we have coolers that work by the thermoelectric effect. But conservation of energy is not the only fundamental law. You have to watch out for things like entropy, so you were wise to ask about other laws.

For example, this one:
 

berkeman

Mentor
55,087
5,316
Summary: Can an air conditioner cool the air in all directions provided it emits energy in some other form?
Block of ice?
 
Last edited:
196
42
Sorry, sounded like you were going to tell me the law it violates, but I got an Amazon link to peltiers. Was that intentional?

My setup sounds pretty sketchy to me intuitively, I have some air conditioner generating 5 units of heat outside 4 inside, and my hope is to just send 2 units of heat elsewhere, when all these things go toward equilibrium. But those peltiers seem to send heat as electricity from cold side to hot side...
 

russ_watters

Mentor
18,578
4,812
I know heat is a form of energy, and that’s conserved, so usually an air conditioner will emit heat on one side, cooling on the other, probably increasing the overall heat in total.

What’s not clear to me is if it always must do this, specifically if the air conditioner produces another form of energy: For instance I’m picturing an electrical device that cools on it’s whole exterior, and uses heat inside to generate output electricity or something through some kind of heat engine on the inside. The Wikipedia page for heat engine says:

“During this process some of the thermal energy is converted into work by exploiting the properties of the working substance.”

What’s unclear to me is if the amount of thermal energy converted to electricity can be large enough that that the net cooling of the device can be greater than the net heating, or if this violates some fundamental laws. (I’m not a physics guy)
Energy is reference frame dependent, so "thermal energy" is a function of the temperature difference between a warm and cold reservoir. No temperature difference means no energy to harness. In fact, for air conditioners there usually is a temperature difference in the wrong direction, which is why they require a mechanical energy input instead of producing a mechanical energy output.
 
196
42
Energy is reference frame dependent, so "thermal energy" is a function of the temperature difference between a warm and cold reservoir. No temperature difference means no energy to harness. In fact, for air conditioners there usually is a temperature difference in the wrong direction, which is why they require a mechanical energy input instead of producing a mechanical energy output.
Indeed, mechanical energy is used to push the refrigerant around to create the cooling. I’m not interested in producing energy, I’m interested in if you can create only cooling in an area by pushing heat out somewhere else on a wire, rather than through e.g. an insulated air hose. My internal heat engine was just a goofy attempt. I’m thinking It can’t be done, but don’t understand just why. Does it follow from entropy?
 

berkeman

Mentor
55,087
5,316
I’m interested in if you can create only cooling in an area by pushing heat out somewhere else on a wire
What does that mean? Conducting heat out on a metal wire, or using electricity to transfer energy? Sorry, I guess I'm not understanding your question so far...
 

russ_watters

Mentor
18,578
4,812
Indeed, mechanical energy is used to push the refrigerant around to create the cooling.
That's over-simplifying: it's thermal energy that you need to create (convert), using the compressor. It's not just a pump, circulating a fluid.
...by pushing heat out somewhere else on a wire...
What does that mean? Typically wires carry electricity, not heat...
I’m not interested in producing energy, I’m interested in if you can create only cooling in an area...
But you are: if you start with a certain amount of working fluid at one temperature and end with it at a different temperature, you've created a certain amount of energy. It had to come from somewhere; you can't just spontaneously create it from nothing.
My internal heat engine was just a goofy attempt. I’m thinking It can’t be done, but don’t understand just why. Does it follow from entropy?
Yes, and depending on the specifics maybe also conservation of energy.
 
196
42
Thanks Russ, bear in mind I’ve never studied physics, I’m just trying to figure what to read online to understand this, having a tough time finding resources on my own, and it’s interesting to me. I’ve already gotten some good hints in your words.

By wire, it could be electrical, but really I’m interested in sending heat energy away in any other form than local heat. Another vision of my picture could be a rock on mars, near a nuclear battery to provide energy. I’m wondering, can I take energy from the battery to cool the rock, while beaming away heat as some kind of thermal radiation into the stars, so the energy achieves only cooling locally? Or does the energy always have to create local heat nearby? I know I can physically move it away from the rock via a fluid, but that’s local heat. I’m interested if it can be made into waves or energy on a wire, like something electrical or EM waves, to create heat elsewhere, and why not.

Just started reading about thermodynamics, but I don’t yet have a holistic view. Any high school/college freshman level resources to wrap my head around heat and energy you know would be appreciated!
 

Tom.G

Science Advisor
2,612
1,443
To remove more heat from the hot side of an air conditioner (the Condenser), you could spray water on it as a fine mist. The evaporation of the water will remove heat from the condenser and carry it away to somewhere else (presumably someewhere else on Earth :smile:).

(yea, not the answer you were looking for, but does answer the question as posed.)

Cheers,
Tom

p.s. Have fun in your quest for knowledge... it turns out to be a never-ending quest!
 
1,280
647
I know heat is a form of energy, and that’s conserved, so usually an air conditioner will emit heat on one side, cooling on the other, probably increasing the overall heat in total.
It is not 'probably' but a law. The overall heat (in a closed system, for example) will be increased by the energy required to run the AC. In an open system the extra heat will be just sent away on the hot end.

Can an air conditioner cool the air in all directions provided it emits energy in some other form?
Well, in special circumstances there is a way. Heat can also be transferred by radiation, depending on the temperature of the source. In normal circumstances it is a near equilibrium since almost everything around is on the same temperature: but if you creates an environment, where the 'outside' is cold (like in space, in a shadow), then the energy radiated by your source will be more than it can receive in return. So it'll work as a heat sink and 'steal' heat from the air.

Of course it is much more complicated than that and requires quite some fiddling to work, so take it just as a poor example...
 
1,345
99
Summary: Can an air conditioner cool the air in all directions provided it emits energy in some other form?

I know heat is a form of energy, and that’s conserved, so usually an air conditioner will emit heat on one side, cooling on the other, probably increasing the overall heat in total.

What’s not clear to me is if it always must do this, specifically if the air conditioner produces another form of energy: For instance I’m picturing an electrical device that cools on it’s whole exterior, and uses heat inside to generate output electricity or something through some kind of heat engine on the inside. The Wikipedia page for heat engine says:

“During this process some of the thermal energy is converted into work by exploiting the properties of the working substance.”

What’s unclear to me is if the amount of thermal energy converted to electricity can be large enough that that the net cooling of the device can be greater than the net heating, or if this violates some fundamental laws. (I’m not a physics guy)

Thanks for any guidance or resources.

If you happen to have some ice and a heat engine in a room, then it's better to use the ice and the heat in the room to run the heat engine, than to let the ice directly suck heat from the room. You get more cooling.

Extra cooling measured in Joules = generated mechanical energy in Joules

If we replace the ice by liquid helium, then the generated mechanical energy may be 10 times the heat energy absorbed by the helium.

If we take your machine, but replace the electric wire by a liquid helium pipe, that machine works almost the same way your machine was intended to work. Your machine was violating the second law of thermodynamics.


But what if we replace the liquid helium pipe by an electric wire and use electric enegy to generate liquid helium? :smile:
 
Last edited:

russ_watters

Mentor
18,578
4,812
By wire, it could be electrical, but really I’m interested in sending heat energy away in any other form than local heat. Another vision of my picture could be a rock on mars, near a nuclear battery to provide energy. I’m wondering, can I take energy from the battery to cool the rock, while beaming away heat as some kind of thermal radiation into the stars, so the energy achieves only cooling locally? Or does the energy always have to create local heat nearby?
Beaming heat away via radiation works and is done, but it does qualify as "creating heat locally". The key is that locally the device creates "hot" and "cold" at the same time. What it does with the "hot" can vary - there are a lot of ways to get rid of it and radiation is one possible way.
Just started reading about thermodynamics, but I don’t yet have a holistic view. Any high school/college freshman level resources to wrap my head around heat and energy you know would be appreciated!
I believe this is the book I learned from (a newer version):
 
Last edited:

256bits

Gold Member
2,748
779
I see a problem with that.
It's not failsafe.
If he inadvertently leaves the unit on all weekend, and the bucket overflows, Monday morning people will come in and there could be 'heat' all over the floor. What a mess.

:H
 

Want to reply to this thread?

"Omnidirectional air conditioner allowed by physics?" You must log in or register to reply here.

Physics Forums Values

We Value Quality
• Topics based on mainstream science
• Proper English grammar and spelling
We Value Civility
• Positive and compassionate attitudes
• Patience while debating
We Value Productivity
• Disciplined to remain on-topic
• Recognition of own weaknesses
• Solo and co-op problem solving
Top