Possible Air Conditioning System?

[mechlite]

Hello everyone!
Summer is coming and instead of buying a $200-$400 air conditioner, I'd figure I could try a DIY version. I came up with a schematic (attached below) and how I think it would work. I feel like I'm missing something blatantly obvious that will make it not feasible. Wondering if you could take a look and give me some pointers?
Essentially:
-Tank of cold water supplies axial fan with cold water to blow via copper tubes.
-Water is sent to a cooling tank that has an enclosure.
-The enclosure contains separate water and an air suction line.
-The air suction line will eventually turn on creating a vacuum.
-The separate water will turn into ice.
-Ice will cool water.
-New cooled water will return to cold water tank.

Please excuse my handwriting. Hopefully it is legible. I suspect the system is not as efficient as an actual air conditioner but it only has to cool 1 bedroom. It is also cheaper this way. Thanks for any input.

 

[berkeman]

Hello everyone!
Summer is coming and instead of buying a $200-$400 air conditioner, I'd figure I could try a DIY version. I came up with a schematic (attached below) and how I think it would work. I feel like I'm missing something blatantly obvious that will make it not feasible. Wondering if you could take a look and give me some pointers?
Essentially:
-Tank of cold water supplies axial fan with cold water to blow via copper tubes.
-Water is sent to a cooling tank that has an enclosure.
-The enclosure contains separate water and an air suction line.
-The air suction line will eventually turn on creating a vacuum.
-The separate water will turn into ice.
-Ice will cool water.
-New cooled water will return to cold water tank.

Please excuse my handwriting. Hopefully it is legible. I suspect the system is not as efficient as an actual air conditioner but it only has to cool 1 bedroom. It is also cheaper this way. Thanks for any input.

It looks like you are trying to use a negative pressure to cool/freeze the water? How does that work?

 

[mechlite]

It looks like you are trying to use a negative pressure to cool/freeze the water? How does that work?

I remember reading that water in a vacuum will begin to boil because of the change in pressure. If the pressure continues to drop or vacuum continues the water will eventually freeze. I think it's sublimation but I can't remember for sure. Is the concept wrong? I never tested it.

 

[berkeman]

I remember reading that water in a vacuum will begin to boil because of the change in pressure. If the pressure continues to drop or vacuum continues the water will eventually freeze. I think it's sublimation but I can't remember for sure. Is the concept wrong? I never tested it.

It says in your Profile page that you are an undergrad ME student. Have you had any courses in thermodynamics yet? If so, can you say what equations and concepts apply to your question and idea?

If not, can you do a little searching and post links to pages that might support your idea of pulling a vacuum in a thermally conductive container to freeze water?

 

[berkeman]

BTW, I'm not trying to give you a hard time. It's good that you are brainstorming and asking questions. But as an undergrad ME student, you should also be doing a lot of reading to send your brainstorming in productive directions.

 

[russ_watters]

Hello everyone!
Summer is coming and instead of buying a $200-$400 air conditioner, I'd figure I could try a DIY version. I came up with a schematic (attached below) and how I think it would work. I feel like I'm missing something blatantly obvious that will make it not feasible. Wondering if you could take a look and give me some pointers?

Thanks for any input.

Welcome to PF!

Um...

It says in your Profile page that you are an undergrad ME student. Have you had any courses in thermodynamics yet? If so, can you say what equations and concepts apply to your question and idea?

If not, can you do a little searching and post links to pages that might support your idea of pulling a vacuum in a thermally conductive container to freeze water?

Yeah. Have you taken a thermodynamics class yet where you have analyzed a refrigerant cycle? What you are describing is a cumbersome version of a standard air conditioner that uses water instead of a typical refrigerant. It has some flaws, but the basic idea would work. What concerns me most though is this:

... I suspect the system is not as efficient as an actual air conditioner but it only has to cool 1 bedroom. It is also cheaper this way.

You haven't done any calculations, but you should at least recognize that "not as efficient" and "cheaper" directly contradict each other: efficiency is what dictates cost to operate.

Anyway, the issue with using water as a refrigerant is that it has a low pressure/high boiling point compared to other refrigerants. That makes it boil really really slowly and require a very large volume to have much cooling capacity. Depending on where you are in your thermodynamics schooling, you should consider attaching actual states to the various steps in the cycle and look at the thermodynamic properties at those states. For example, after your evaporator, what should the temperature and pressure of the water be?

Anyway, you might be jumping the gun here a bit if you haven't had your first thermo class yet, but don't get discouraged: the thoughts and willingness to experiment on your own bode well for your potential future as a mechanical engineer. My condensing unit still has micro-sprinklers attached to it as a remnant of a minor success of an experiment with supplemental evaporative cooling...

 

[mechlite]

I should clarify. I am going into engineering. I am not a grad so I have not taken any thermo classes.
I saw the video below and wondered if it would work. Figured I'd ask on here to get some more guidance/clarification. When I said less efficient but cheaper I meant it won't cool the room as much as an actual ac but at least it will cool it down a little and be less expensive. I apologize for the confusion and posting without doing calculations.

https://van.physics.illinois.edu/qa/listing.php?id=1723

 

[russ_watters]

I should clarify. I am going into engineering. I am not a grad so I have not taken any thermo classes.
I saw the video below and wondered if it would work. Figured I'd ask on here to get some more guidance/clarification.

Ok, well as I said, yes it will work. It is just a cumbersome version of how a standard air conditioner works. You'll learn all about it in thermo, but the wiki article should get you started:
https://en.wikipedia.org/wiki/Vapor-compression_refrigeration

When I said less efficient but cheaper I meant it won't cool the room as much as an actual ac but at least it will cool it down a little and be less expensive. I apologize for the confusion and posting without doing calculations.

That's ok, but it still doesn't compute: if the device is cheaper to run because it doesn't do as much cooling, then just changing the thermostat setting to match on a standard AC unit would save even more money. The non-apples-to-apples comparison just makes the lack of benefit harder to see.

 

[Mech_Engineer]

Vacuum pumps can be expensive and might not last long if they're pumping a lot of water vapor. Are you sure this idea is less expensive than a more conventional refrigerant-based approach?

 

[mechlite]

Ok, well as I said, yes it will work. It is just a cumbersome version of how a standard air conditioner works. You'll learn all about it in thermo, but the wiki article should get you started:
https://en.wikipedia.org/wiki/Vapor-compression_refrigeration

That's ok, but it still doesn't compute: if the device is cheaper to run because it doesn't do as much cooling, then just changing the thermostat setting to match on a standard AC unit would save even more money. The non-apples-to-apples comparison just makes the lack of benefit harder to see.

So it works on paper but doesn't translate in real life application. Fair enough. I suspected I had missed something blatantly obvious, but didn't see it. Maybe with more education I can have the foresight to see why things work/don't work. Additionally, save myself some embarrassment due to ignorance. Thanks for the help and I will look into refrigeration cycles.

 

[mechlite]

Vacuum pumps can be expensive and might not last long if they're pumping a lot of water vapor. Are you sure this idea is less expensive than a more conventional refrigerant-based approach?

I initially thought it would but I think using a conventional refrigerant would be better. I have access to a vacuum pump and I saw that video a while back and thought of that system. Kinda glad I asked before buying the other stuff.

 

[russ_watters]

Additionally, save myself some embarrassment due to ignorance. Thanks for the help and I will look into refrigeration cycles.

Ignorance should not cause embarrassment at your current level; accepting it is the first step to learning (you wouldn't have asked the question otherwise!).

 

[jim hardy]

So it works on paper but doesn't translate in real life application. Fair enough.

The U S Navy used just that system on some of its submarines. Their vacuum pump is a steam powered "Eductor", basically enough venturis in cascade to move the required water vapor out of the evaporation chamber to ? (i'll guess probably the ship's condenser ?)..
With no rotating machinery it can be made near silent an obvious advantage for submarine...
Of course they had plenty of steam available to run it...

I think it makes sense for a solar powered residential airconditioner booster in the Sun Belt.
If you evaporate fifty gallons of water over the course of a day you have provided about a ton of refrigeration.
In a batch machine the size of the components is manageable. We always think in terms of a continuous process but solar is inherently batch mode.

So keep on with your investigations.

When i lived in Florida i really wanted to build one. The collectors would shade the roof and an automobile radiator mounted in my air-conditioning plenum would help the Freon system cool the house.

Start putting numbers on your design.

old jim