Join 3,503 readers in helping fund MetaFilter (Hide)


Woah, cool!
June 15, 2005 7:14 AM   Subscribe

Roll your own air-conditioning. I'm a student, with limited funds and a cheap house without air conditioning. To avoid dying this summer, I've built a primitive air conditioner. It's a basic heat pump, using water as the medium. You'll probably need to fiddle a bit with the dimensions of the supplies based on your resources and preferences. Not sure I'd do this but hey, when you're sweltering hot, anything is worth a try.
posted by KevinSkomsvold (48 comments total)

 
The condesation coming of the back copper coil is what gives me pause. I see electrical issues with that.
posted by KevinSkomsvold at 7:15 AM on June 15, 2005


I see issues with having to refill a garbage can with ice water every couple of hours.....

But, that setup does add to the decor of the room darn it...
posted by HuronBob at 7:20 AM on June 15, 2005


It ties the room together, don't it? Yeah, might wanna toss some wicker or drapes on that thing.
posted by KevinSkomsvold at 7:22 AM on June 15, 2005


Or just pick up a Swampy for about $10 more.
posted by [insert clever name here] at 7:22 AM on June 15, 2005


Good for him. Basically, he's created a "poor man's" DeLonghi Pinguino air-conditioner.
posted by ericb at 7:23 AM on June 15, 2005


While the homebrew a/c is interesting, Mirrordot is really interesting. A little websearching even led to a "Business Week" story that features our own mathowie. Neato.

Thanks, KevinSkomsvold!
posted by Stoatfarm at 7:31 AM on June 15, 2005


Yeah, I saw this on Fark, and agree with the general comments found there: this is not true AC, but merely blowing air over ice (essentially). It is also very wasteful of water.

You would generate more heat from making the ice you have to put in the thing than you could remove via this method.

However, I must admit that, had I thought of/known about this back in my dorm days (where the ice machine was in a different room and made basically unlimited ice) before I cared too much about wasting water, I would have probably done it.
posted by Yellowbeard at 7:40 AM on June 15, 2005


If you had ice to begin with... wouldn't it be more efficient to...

nevermind.
posted by odinsdream at 7:49 AM on June 15, 2005


I think it's pretty cool, but with all of these kinds of things, I suspect that the thinking about it/building it/fiddling with it/writing about it was as much a factor as the getting cooler. I'm not saying he did it for attention, I'm saying he did it to have something interesting to do.
posted by OmieWise at 7:50 AM on June 15, 2005


Heck, we used to hang a towel dampened in cold water from the tap across the back of a box fan. I suppose this way controls for humidity better though.
posted by Fezboy! at 7:53 AM on June 15, 2005


He's going to spend more on icecubes and electricity then he would have if he'd gotten a new AC unit.
posted by delmoi at 7:56 AM on June 15, 2005


You would generate more heat from making the ice you have to put in the thing than you could remove via this method.

Of course, if he put his fridge out in his back yard or balcony, or whatever...
posted by delmoi at 8:00 AM on June 15, 2005


I've been wondering when someone was going to come up with a homebrew AC site.

I've been thinking about grabbing one of the old computer cases scattered around my closet and using it as the framework for an Evaporative Cooler (or Swampy as some would say).
posted by chibikeandy at 8:00 AM on June 15, 2005


Wouldn't Mister Fix-it here be better off removing his refrigerator's condenser (extending it with a little extra tubing) and motor (add longer wires) and moving them outside or into a cool basement? If the fridge is near a window (or if he doesn't mind drilling a couple of little holes in the wall behind the fridge), it wouldn't have to be much of an extension.
posted by pracowity at 8:07 AM on June 15, 2005


What the hell? It gets hot in Canada? Go spend a week somewhere where it actually gets hot and you'll go home and never turn on your AC again.
posted by loquacious at 8:12 AM on June 15, 2005


Does my freezer work harder when it is full of water? As I (like most people) keep my fridge on 24/7 could I generate a steady supply of ice utilizing power I'm already paying for? Or do I use more juice when I toss in that extra water mass?
posted by sourwookie at 8:18 AM on June 15, 2005


What the hell? It gets hot in Canada? Go spend a week somewhere where it actually gets hot and you'll go home and never turn on your AC again.

Canada is a big, varied place. Where I grew up, in Merritt BC, it would reach 40 Celsius (about 103 Fahrenheit) for much of the summer. That area is desert...cactus, rattlesnakes and all.
posted by Kickstart70 at 8:40 AM on June 15, 2005


They used to make installed units that pumped city water like this, a friend of mine still has one in Toronto. It is pretty much not allowed anymore around here, although I don't know exactly what statute stops it (is the device banned for sale, is it against your water use agreement, whatever).


sourwookie:

Your freezer works harder if you keep loading it with warm stuff. The work it does is directly proportional to the amount of energy it has to move from inside the freezer cavity to the rest of the room. Since there is a lot more energy in a warm glass of water than the same volume of air...

If you don't keep reloading it with warm stuff having it full makes it work less. This is because when you open the door of an empty freezer a lot of warm air gets in, not so if it is full of stuff.
posted by Chuckles at 8:41 AM on June 15, 2005


Does my freezer work harder when it is full of water? As I (like most people) keep my fridge on 24/7 could I generate a steady supply of ice utilizing power I'm already paying for? Or do I use more juice when I toss in that extra water mass?

Your freezer does work harder. think about it this way: Imagine a row boat that has a hole in it. As long as you keep bailing, you can stay afloat. The bigger the hole, the faster you have to bail. Now, if someone comes along and dumps a bucket or two of water in there from outside, you will just have to bail faster to catch back up - right?

Well, your freezer is like a boat of cold. It must work to keep the heat out. Insulation helps slow the heat's intrusion (like the wooden planks of your boat help slow the intrusion of water) but it can't stop it (like the leak in your boat). Heat is always seeping in, and your freezer is constantly "bailing" by circulating cold freon into the coils - but the heat from that freon must go somewhere (this is slightly simplified, but basically accurate). Just touch the copper tubing on the back of your fridge and you will see where the heat goes. Unfortunately, it costs more energy to keep the heat out than the amount of heat removed from the inside of the refrigerator (see the third? law of thermodynamics). That is, the amount of temperature drop inside the freezer is not as much as the increase in temperature outside the freezer - you are creating extra waste heat.

Now, putting in warm water (directly introducing heat into the inside of the freezer box), not to mention just opening the door, is like dumping buckets of water into our previously mentioned row boat.

Does that make sense/answer your question?
posted by Yellowbeard at 8:45 AM on June 15, 2005


Go spend a week somewhere where it actually gets hot and you'll go home and never turn on your AC again.

The problem with growing up where it's cold during the winter(not necc. Canada) is that when it first gets hot, it feels really really hot (at least at first). On the plus side, we can wear shorts in 50 degree weather.
posted by drezdn at 8:54 AM on June 15, 2005


< farksnark>Thermodynamics surrenders.< /farksnark>

It would have been more efficient to just leave the freezer door open and put a fan in front of it. It's a dorm room, right?

Also: What loquacious said, succinctly.

He should consider himself lucky. Swamp coolers don't work so well in the 90% humidity of the Southeastern States.


posted by Mr. Gunn at 9:01 AM on June 15, 2005


What the hell? It gets hot in Canada? Go spend a week somewhere where it actually gets hot and you'll go home and never turn on your AC again.

What Kickstart70 said. Plus, when your house is designed to retain heat during the long winter months...well, you can see how that might be detrimental when it does get warm. My house takes a while to heat up when it's hot outside, but once it does, getting it back down to a bearable temperature is a bitch.
posted by The Card Cheat at 9:18 AM on June 15, 2005


Meant to include: what drezdn said as well.
posted by The Card Cheat at 9:19 AM on June 15, 2005


"I got the idea when I noticed the refrigerator was cold."
posted by Otis at 9:24 AM on June 15, 2005


I have a large five room apt. cooled by a single window unit. When I saw the Swampy I thought "Hey! A little unit I can take to my bedroom at night!"

Unless, of course, I would significantly increase my utility bills by making an arctic flow's worth of ice every day.
posted by sourwookie at 9:33 AM on June 15, 2005


I'm wondering if there's anyone who lives in the basement, and if they appreciate the endless stream of water draining out in front of their door and probably seeping into their apartment.


posted by chococat at 9:38 AM on June 15, 2005


Instead of just draining it onto the ground, he could drain it into a second garbage can for reuse and cut down on the water waste.
posted by fandango_matt at 9:59 AM on June 15, 2005


For what it's worth, I have a plan (in my head) for an evaporative air cooler. Basically it's a box with two compartments. The compartments are joined by a hole toward the top of the separating wall. There is a similar hole at each end of the box. These 3 holes allow air to flow from one compartment to the next and out the other side of the box.

In the first compartment, pour some water, filled to below the level of the holes in the walls (of course). Hang some strips of muslin from the lid of the box into the ice water to create a larger surface area for the water, so it evaporates quicker. In the second compartment, add a load of dessicant granules, to hopefully extract some of the moisture from the now-cooled air. Point a desk fan toward the hole at one end of the box.

Now, if my thought-experiment works, it ought to produce cool air :-)

I haven't actually built it yet, but my naive and optimistic brain is convinced it must work.

I agree with all the other posts about the wastefulness of the system in the article. Air-conditioning (or rather, air cooling in this case) is hugely inefficient, and this homebrew version seems to be even more so than a commerical unit, although the basic ideas are identical.
posted by ajp at 10:02 AM on June 15, 2005


ajp: Evaporative air coolers are common in the form of "cooling towers", and are used frequently in large A/C installations.

The open-loop air/water system you describe (basically an a/c without the coolant) would work very well in hot dry areas (They're used extensively in the mideast iirc). Unfortunately, in hot-dry areas water is often too expensive to pump continuously through an evaporator. In hot areas where water is plentiful, the air is usually saturated with humidity already, making the evaporator, well, evaporate less.
posted by Popular Ethics at 10:18 AM on June 15, 2005


The gravity drain is a nice feature, since it doesn't involve any moving parts or extra gear--it just works, simply. However, it also wastes water, and purchase of a small electric water pump would allow pumping water back into the reservoir rather than out the window. Thus a smaller reservoir could be used, also. Maybe like the $40 Swamp Cooler linked above, I don't know.

I was hoping when I first clicked on the link that the guy had built his own real air conditioner--those use mechanical expansion and compression of a coolant to operate, much like the can of hairspray getting cold when it's discharged, only in a continuing cycle. The linked contraption is described as a primitive Heat Pump, but as far as I know, a Heat Pump also uses a mechanical cycle with expansion/compression of a coolant.
posted by nervousfritz at 10:20 AM on June 15, 2005


Yellowbeard writes "Yeah, I saw this on Fark, and agree with the general comments found there: this is not true AC, but merely blowing air over ice (essentially). It is also very wasteful of water."

This is the very definition of AC: Blowing air over something cold and continuously replacing the cold. His concept isn't new either it is how water cooled condensers work. The big advantages: lower head pressure (water is colder than ambient air) and you don't need an outside area to install a condensing unit (Like a ground floor coffee shop in high rise). Toronto has a big project where they extended their water intake system into deep water. They then pump this directly to cooling towers in high rises where it improves efficiency and warms the water to something that is use able by houses.

Also bring to mind 19th century ice refers though they also had an evaporative effect going for them.

Mr. Gunn writes "He should consider himself lucky. Swamp coolers don't work so well in the 90% humidity of the Southeastern States."

This isn't a swamp cooler and would have no effect on RH in the room if he covered that big vat.

Chuckles writes "They used to make installed units that pumped city water like this, a friend of mine still has one in Toronto. It is pretty much not allowed anymore around here, although I don't know exactly what statute stops it"

They probably just started to meter water. Interesting point: it is possible to effectively steal (if they pump it to generate head pressure) electricity from the water utility if you're not metered. What you do is install the head end of a turbine into your water line (preferably upstream of your pressure reducer) with the waste going to your sewer. You can get a significant amount of power 24X7 on even 40psi head pressure and a 3/4" inlet.
posted by Mitheral at 10:31 AM on June 15, 2005


ajp, if you live in an environment of low humidity, then evaporative cooling is quite efficient. The unfortunate thing is it doesn't work in high humidity. Those of us who live in dry-booger states welcome the added moisture in the cool air a swamp cooler provides, and don't need a dessicant stage. Imagine the energy and heat produced to dry the dessicant when it becomes saturated.
posted by Eekacat at 10:35 AM on June 15, 2005


How effective do you all think it would be to strip the cooling coils from a mini-freezer (or mini-fridge, even) and attach them to the back of a box-fan?
posted by Jon-o at 11:08 AM on June 15, 2005


Jon-O, price wise it would cost roughly as much as a cheap a/c.
posted by drezdn at 11:18 AM on June 15, 2005


How effective do you all think it would be to strip the cooling coils from a mini-freezer (or mini-fridge, even) and attach them to the back of a box-fan?

Really effective, if you have a way to continuously pump liquid nitrogen through them. Well, then again, the back of the box fan will probably ice over, preventing air flow through the coils. So you'll have to probably have a dehumidifier inline with the boxfan some how. Good luck!
posted by nervousfritz at 11:20 AM on June 15, 2005


Well, I'm thinking that not only do you remove the coils from the fridge, but essentially all of the components that cause it to function properly. Really, everything but the fridge enclosure.

My friend doesn't have an AC and he's TOTALLY MISERABLE. I was thinking of this as a favor or as a surprise for him...
posted by Jon-o at 11:25 AM on June 15, 2005


Back when I was in the dorms in Iowa, we'd tie those blue freezer pack thingers to a fan, and swapped them out with other icepacks sitting in the hall freezer.

Condensation was indeed a problem, but it probably helped minutely reduce the humidity of the room.

/we also thought of criminally wasting water by piping it from a faucet outside and into the dorm, wrap it around the fan, and just drain the bugger out the window
posted by PurplePorpoise at 11:29 AM on June 15, 2005


You know, you can get an air conditioner at Home Depot for $79.
posted by MrMoonPie at 11:33 AM on June 15, 2005


You know, you can get an air conditioner at Home Depot for $79.

Pffft. Where's the fun in that?
posted by deborah at 12:41 PM on June 15, 2005


My air conditioner goes to 11.
posted by Debaser626 at 1:04 PM on June 15, 2005


Refrigerators are designed for a lower coil and head temperatures than A/Cs[1]. Converting one as proposed would be both horribly inefficient electricity wise (though better than this rube goldberg device) and also prone to freeze up. What you want in compression refrigeration systems is for the condensing point temperature to be at or slightly below your output design temperature. Too much variation up or down and it costs you efficiency. There are literally hundreds of different refrigerants, all with different characteristics, for different uses.

[1] For example you fridge probably never sees high side temps of much more than 150-180F with ambient around 75 and coil temps down around 32 to -10F depending on application. I've work on roof top units where the ambient air temp was 150F and high side temps were pushing 300F with coil temps around 45-55F. Sorry for all the farenhiet temps, even though Canada is metric I apprenticed under an old school guy and I think about refrigeration in F.
posted by Mitheral at 1:23 PM on June 15, 2005


I've been to the Caribbean and Africa, and I still find North Bay Ontario to be the hottest and most uncomfortable place I've ever been in. Go spend a week there and you might actually understand...
posted by Elpoca at 2:27 PM on June 15, 2005


I'm moving to Tucson, AZ, and you'd be amazed at how many people who live in the freakin desert do NOT have A/C, but just have these evaporative "swamp" coolers. Seems insane to me here in Texas, where you don't survive without A/C. But, that said, they seem to work OK, except for the 1-2 months that it rains.
posted by papakwanz at 2:30 PM on June 15, 2005


Evaporative coolers can drop the temps a good 40-50F in conditions of low RH. And a shaded 40C at 15%RH feels a lot better to me than 30C and 98%RH. You don't get much from sweating when the RH gets high.
posted by Mitheral at 2:58 PM on June 15, 2005


What the hell? It gets hot in Canada? Go spend a week somewhere where it actually gets hot and you'll go home and never turn on your AC again

It's been hovering around 30 degrees C (about 86 degrees F) in Toronto for the last week and a half. And the humidity makes it feel about 10 degrees higher. Trust me, in the summer, it's hot.

Toronto's weather is much like New York's.
posted by jrochest at 3:39 PM on June 15, 2005


papakwanz, I live in Tucson, AZ, and my house is cooled by a swamp cooler. It works great, and even does fairly well during the month-long rainy period; it's not like it gets *that* humid here even then.

I would not use one in a humid climate, however. It wouldn't work. But it's great here in the desert. I actually like it better than A/C.
posted by kyrademon at 9:49 PM on June 15, 2005


"It's been hovering around 30 degrees C (about 86 degrees F) in Toronto for the last week and a half. And the humidity makes it feel about 10 degrees higher. Trust me, in the summer, it's hot."

Dude. 86 is winter weather in texas.
posted by muppetboy at 10:36 PM on June 15, 2005


Yeah, swamp coolers don't work well here next to the ocean in Vancouver. All it does is get the things in front of it wet.

Up in the dry southern BC interior, they are awesome though.
posted by Kickstart70 at 1:00 PM on June 16, 2005


« Older In Congo, 1,000 die per day: Why isn't it a media ...  |  Keep your ice cream safe... Newer »


This thread has been archived and is closed to new comments