An analogy would be if you had a window AC sitting on a table, with the hot fins surrounded by a box. In the box, you put a bunch of towels. You pour some water onto the hot fins to cool them down, and then close the box. The water quickly evaporates, and gets soaked up by the towels. You bring them outside and squeeze them out. You then have a bucket full of hot water, which you let cool down outside for awhile. Then you dump the water back on the coils and repeat.
The analogy is crude, because a lot of the water would evaporate to the outside air, etc. But it might aid in understanding that business I described about adding a second water compressor.
For me, there are two problems with portable unit. First, the one I have is supposedly a 10,000 BTU unit, but it doesn't cool even a very small room on an 85 degree day. Maybe a degree or two temperature drop, if that. The coils aren't dirty, it's charged just fine, and there aren't any leaks. The area right in front of the unit is cold, but an equal amount of heat must "leak" back into the room. I've used it in different locations, with the same results. Most of the problem probably comes from the (very) hot exhaust hose. I'm considering adding some insulation to it. Some heat might also come from the chassis of the unit itself. Even if I had the ambition to built a better fiberglass chassis and put insulation inside it, I think doing so would cause the unit to overheat.
And the second problem is just that the hose is so large and ugly. And since it's so hot, the unit has to be RIGHT next to a window. And if the unit has to be right next to a window, and it only cools down the area immediately in front of it, it pretty much limits the places you can work.
Maybe the solution is to just assume that the units are 1/2 to 1/3 their stated BTU rating (when comparing to a window unit), and buy accordingly, then insulate the exhaust duct. But most of the consumer-oriented portable units only go up to 10,000-12,000 BTU (which SHOULD be fine for a small room). If you want higher than that, you have to go to the ones geared towards wharehouses and computer server rooms. They're extremely expensive and you wouldn't want them in your living room, even if you had the proper power hookups. So, unless insulating the duct makes a WORLD of difference, I'm just going to assume that portable AC's aren't for me - they're more for emergency cooling and spot cooling.
We do pay for hot water. I agree that if I pushed the water back into the hot water pipe, it would probably be used up. But wouldn't doing so require spending a lot of energy to force the water against the pressure in the pipe? Or would you rely on thermosyphoning alone? I do agree that it would be a shame to miss the energy-savings of heating hot water, but I think I can wait until my home is permenantly setup before attempting that part.
-------------
As far as evaporating the hot water, having the 2nd compressor, etc:
Since it would be a closed loop, any mineral deposits would flush out once, and stay out (it will "auto-distill"). Distilled water could just be used from the start. Corrosion of the hot fins on the AC is definitely something worthy of consideration. Maybe something other than water (but still not freon) could be used.
I think this type of system would be less efficient than a regular window AC, and you wouldn't get the benefit of the hot water usage. But I think it would still be more efficient than a portable AC, and about equal in efficiency to the pool setup. If it was setup precisely (ie water flow rate, pressures in different parts of the circuit), the water compressor would barely be doing anything. It would just be providing a "nudge" for the water vapor to go liquid again, and the outside air would do the rest. And because the hot fins of the AC would (hopefully) be MUCH cooler than they would be if the unit was mounted in a window, the AC wouldn't have to work as hard (ie: less energy spent).
---------------------
Can you explain further about the plastic films? What type of plastic would be used, and what are its thermal properties? I'm not familiar with that type of thing.
Please explain what you do not like about the portable unit. I have just purchased one for my bedroom and am now in the process of building a 2' high pedestal so the exhaust hose can get a straight shot out the window. (I love the concept of the portable unit because it is one less thing I have to deal with to prepare the house for a hurricane.)
I've never run out of hot water in this place. It's split equally.
So the water comes in from the cold pipe under pressure, and hopefully doesn't lose much pressure throughout the circuit. Then it goes through this small pump and gets forced into the hot water pipe, which is at an almost equal pressure?
The one problem with this is that if the hot water tank *does* fill up, I don't have access to it. I will just have to dump the hot water directly down the drain and take more cold water. What was the gallon per minute flow rate you quoted early (ballpark figure is fine)? Take that flow rate and compare it to the average hot water usage for a few apartment residences. During showers or dishwashing, the two flow rates might be comparable. But otherwise I think the hot water tank would fill up very fast. Filling the hot water tank quickly is good, in one sense, but afterwards I will probably be back to dumping nearly a thousand pounds of water per hour down the drain pipe.
I could COMBINE that idea with the outdoor cooling circuit idea. But it wouldn't be as simple as just redirecting the heated water outdoors. I would also have to stop taking in cold water, and take in the water which is returning from the outdoor cooling circuit. It would require four solenoid valves, or two 3-way solenoid junctions. Now that's fine. But the REAL disadvantage for me is that there are now two constraints:
- Must have piping going to the hot and cold water pipes
- Must have piping going outdoors
So now the AC is somewhere near a power outlet, and there are 4 rather long tubes coming out of it. At any given time, water is moving through only two of the tubes - one hot and one cold. It would work fine in a basement, but it's a little much for an apartment (at least mine). In my apartment, one side faces the outdoors (porch & windows on this side) and the plumbing runs on the opposite side (kitchen & bathroom on this side).
I'd be fine with putting the AC on either side. Meaning I'd use either the outdoor cooling circuit idea or the hot water pipe idea. But I'd rather not do both. For me, I think I'd prefer the outdoor cooling circuit, just so that I didn't have to dump all that water down the drain. I know, it's free for me, and I was considering it before. But I've recently come to think that it is rather wasteful of resources. Also, they might notice around 2 gpm * 1440 minutes per day = 2880 gallons per day of additional cold water usage.
airflow,
Well, here's what I was planning. The water would start off as steam, then pass through the compressor, raising its boiling point (turns liquid at a lower temperature). It would go through a condenser (basically a radiator), where it would cool and turn liquid. It would then head towards the hot fins of the indoor AC. When it got there, the water tubing would enter a chamber which is hot (from the AC fins) and is at a lower pressure (because the water would be expanding into a larger volume). You're right that it would require a vacuum-tight seal. But the fins wouldn't be underwater, and no outside air would pass through them.
BTW: To avoid corrosion, maybe use glycol mixed with an anti-corrosive agent instead of water.
I'm not sure. Some tight rubber gaskets around the refrigerant tubing might work. Without a vacuum-tight seal, this evaporative water circuit couldn't be used (at least not *directly*).
Awhile ago, I described an *indirect* approach for transferring heat between two closed circuits (the liquid-liquid heat exchanger). But I've since decided that is probably a poor method (performance degredation).
Maybe instead of rubber gaskets, the chamber could be built around the fins with a small clearance wherever the refrigerant tubing got in the way. Then I could weld around all the clearances.
The heat would dissapate from the tube into the air naturally; the water (vapor) is hotter than the air. A standard radiator/condenser would do the trick. A fan would help.
"water circuit" air conditioner water tower air conditioning
I found information on two types of AC systems which use water circuits to transfer heat. The first was used mostly in large boats / yachts. These people use closed circuit fresh-water circuits to directly distribute heat to / absorb heat from different parts of their yachts. One such system is described at the following web page:
formatting link
On that web page, they describe three types of systems:
1) Direct expansion, single room unit. Pretty much a standard window AC made for marine use.
2) Central compressor, direct expansion. Basically a multi-split.
3) "central systems with chilled (or heated) water distribution to several fan coils "
The THIRD system is the one I am interested in. It would be equivalent to putting almost everything AC-related outdoors. A water circuit would just come indoors, and pass through a fan-assisted radiator. A small radiator is capable of doing the job; I know of people who cool large areas (sucessfully!) by passing well water through a truck radiator. The water just has to be kept cold enough (well-water temperature is fine). Anyway, onto the second type of system.
The second type of system relies on a cooling tower, and is often found in large commercial buildings or hotels. I've actually seen these before and talked to some people who service them, so I don't know why I didn't remember them earlier. It's fairly similar to what we were talking about with the cooling circuit on the porch, though there are some differences. Here are the basic principles, which I've just copied and pasted from another site:
- A cooling tower blows air through a stream of water so that some of the water evaporates.
- Generally, the water trickles through a thick sheet of open plastic mesh.
- Air blows through the mesh at right angles to the water flow.
- The evaporation cools the stream of water.
- Because some of the water is lost to evaporation, the cooling tower constantly adds water to the system to make up the difference.
That is actually just one type. There are both direct (open circuit) and indirect (closed circuit) versions. A good source of basic information about them is:
formatting link
With an direct (open circuit) cooling system, only regular water can be used in the circuit, because it will constantly be evaporating and being replaced. So the fins would probably need to be plated with something which is corrosion-resistant. With an indirect (closed circuit) cooling system, the main cooling circuit doesn't have to be filled with water, and an anti-corrosion additive could be used.
Now, the "usual" purpose of these cooling towers is to increase efficiency. But for me, it has the side-effect of letting me place the AC indoors, with the hot fins being cooled by the water stream. I have a feeling that THIS system would be best in efficiency (rather than the other system - used on boats). The water might be around 90 - 110 degrees (20 or so degrees above outdoor temp), and the fins might be around 190 degrees. The difference in temperature between the water and the hot fins would be rather large, so heat transfer would be rapid. With the boat system described earlier, the cold water might be 60 - 70 degrees. The room temperature would only be about 10 - 20 degrees hotter than this. So heat transfer wouldn't be as rapid.
The one advantage of the "boat system" is that the entire AC unit could be placed outdoors. On the indoor side, there would be no tubing above room temperature. Which eliminates the problem I have had with my portable unit: heat leaking back into the room. As far as I'm concerned, both systems are adequate; ease/difficulty in construction will be the deciding factor.
Well, previously I said that I would just buy a system if I could find a water-based system which was of about the size I need.
formatting link
Starting price: $1,189 for a 16,500 BTU unit that would go indoors, and be water-cooled to an outdoor "cooling tower" (this I know I can make).
formatting link
Starting price: $3,300 for a 18,000 BTU unit that would be almost completely outdoors, except a cold water circuit and a fan-assisted indoor radiator. Since it is intended for marine use, it is probably water-cooled, so an outdoor "cooling tower" might still be needed.
The 1st one is cheaper, and probably more efficient, so I might just buy that. I could either hide it in a chest / cabinet (with an air grille), or make a fiberglass chassis for it.
Short of moving, or cutting a hole in the wall (and getting evicted), what would YOU do? (I'm expecting some not-very-funny comment about working in the car/bath tub... or possibly something more childish like "die")
Back when I purchased my portable AC, it was somewhere near $1000 (slightly less). Even if it worked a little better, it will never be able to cool more than a bedroom; it's only about 10,000 BTU. Portable units don't come much bigger than that without spending absurd amounts of money.
You still haven't posted a single intellegent comment, of either a technical or theoretical nature, which backs up anything you've said.
Good. Sounds like you have one big water heater tank...
Yes.
Maybe. You might just keep pumping hot water back into the pipe.
About 5KBtu/h/(110F-60F)/8.33lb/g/60m/h = 0.2 gpm.
You might use 50K Btu/day of AC, roughly the hot water usage in one apartment. With more than one apartment, you might seldom if ever need to dump any hot water. Skip that feature for starters. Worst-case, people find warm water coming out of their cold water taps, and let the cold water run a bit longer.
If nobody (including you) uses any hot water, you might eventually have to dump 0.2 gpm (about 100 lb/h), but that's unlikely. Consider learning _a tiny smattering_ about heatflow :-)
Yes, that's possible, but like you say later in this post -- the cold water is going to start getting heated, if it even works (assuming there's no 1-way valve in the water heater, or anything like that).
And when you have a partially full water tank, you have a resevoir to pump the heated water into. If that tank is full, and you continue to try to force the contents of the tank back into the cold water line, aren't you depending on the (near) neighbors using their COLD water to allow this to happen? It's virtually incompresible, and it has to go somewhere. I mean, in theory, you could force the water further back towards the city's water plant. But if it's not used in the immediate area, you're going to need a much bigger pump to push the water - due mostly to its viscocity. I don't think that's the way to go.
Not that putting the hot water back into the heater is a bad idea -- you just need a secondary option once the heater fills up.
Ok. I should have done a rough calculation, that's not all that bad.
There's another problem, even if the neighbors didn't notice or care. If the cold water pipe started getting filled with hot water, and the AC circuit was USING that same cold water pipe to cool itself..... Granted some other people would also be using both the hot and cold water occasionally... But essentially you'd have a quasi-closed loop - the cold water pipe - cooling the AC. Problem is, after awhile, the water in the "cold water pipe" wouldn't be so cold. It all depends on everybody else's usage.
Yea, sorry, I just pulled a figure from a previous post without checking it. The figure I quoted was the water needed to provide cooling simply by passing the water through a radiator -- the original idea I posted about.
Hmm Paul, that's funny. It seems that after doing some research, none of the ideas I've suggested were even new ideas. Now that part is my fault; I should have done more research before posting. But not only are they not new ideas, guess what: they've ALL been DONE before, and work. And are sold off-the-shelf. Now don't try to back out and say that you only meant that I couldn't BUILD them - you definitely thought that some of the very concepts just COULDN'T work.
Go work retail.
Ideas that have been suggested:
Cold well/tap water through radiator
- Been done, works (search google and google groups, some guy has a web page on it, and LOTS of people do this with lakes or rivers)
- People even bury PVC pipes under the soil and just run air through the pipes with a simple fan. It works. Same concept. (again, search google)
Cooling hot part of AC with a water stream
- Been done, works (boat AC's, commercial AC's with water towers)
Cooling a stream of water using an air conditioner (ie: chiller) and then passing the cold stream through an indoor radiator with a fan
- Been done, works (boat AC's)
Cutting the refrigerant circuit of a regular AC, then putting the evaporator indoors:
- Been done, works ("multi-split" ACs)
- As far as building, need a vacuum pump, freon draining equipment, soldering equipment, etc (not as hard/expensive as you make it sound)
Nick's hot water tank idea:
- Been done, works (Counterflow heat exchanger)
-
formatting link
His idea is implemented a little differently, but the results are similar
The exact AC unit I've been describing recently, and wanted to build: Portable, non-marine, 20,000+ BTU, water cooled AC
- Been done, works
-
formatting link
formatting link
MANY more companies produce them
- Water consumption for a 10,000 BTU unit:
- .6 GPM @ 60°F Inlet/ 2.5 GPM @ 85°F Inlet
Outdoor water cooling "tower"
- Can easily be built with plastic sheets, as Nick described
- Even though usually found on large AC's, can easily be built for one of the smaller water-cooled AC's above
Also, these water-cooled or water-cooling units that we are suggesting have ANOTHER benefit that neither a window AC or a portable AC have: there is no negative pressure created inside. Hot air does not get pulled in from the outdoors through cracks in the windows, etc (maybe due to diffusion, if there are huge cracks, but that can usually be ignored). That is what one of the other posters was mentioning when we were talking about the poor performance of a portable AC I was using.
It turns out that EVERY single one of these ideas not only CAN work, but is available off-the-shelf for relatively little money (a little more than a common air-cooled portable AC). So Paul, you're dead wrong, on all counts.
As far as comments about BUILD> You might also glance at some basic fluid dynamics books,
Not if you use less AC energy than the water heating energy used by you and your neighbors, which seems likely.
But the hot water tank is always full.
No.
It goes back into the hot output of the water heater and pushes cold water out the cold input, which comes out of the cold water pipe in your apartment and goes into the heat exchanger (the copper pipes inside that 8' tall PVC pipe) and gets heated and goes back into the hot water pipe. It's a loop. Nothing explodes :-)
In the _unlikely event_ that that happens, you either dump some hot water into the drain in your apartment or you and your neighbors let the cold water run a bit more until it gets colder.
Let's stop now. It's clear you don't have enough knowledge or inclination to do this. Maybe someone else will :-)
Actually, that may be part of your problem. If your AC is blowing hundreds of cubic feet of hot air out of your room, then an equal amount of air must enter the room from *somewhere*. I have seen a few units that have two hoses, but that may lead to other problems. Also, just because the units come with 7 or 8 feet of hose, it does not mean that it is a good idea to use it all. My unit will have about 18", of hose that will exhaust out in a straight line.
I doubt that is a significant part of your problem. Check with me in three or four weeks and I will have some practical experience with my new unit. Actually, my house has central air, but since we are "empty nesters" we can get away with shutting it off at night and just running a small bedroom unit. Also, our standby generator can power that bedroom unit in the "powerless" days following a hurricane, but is not sufficient to power our central air.
We both agree on that - I'm not trying to say it couldn't work - I really do like the idea. It's just a matter of guessing whether or not that condition holds.
Oops, my mistake. That makes things worse though.
happen?
If nobody else was using cold or hot water, and the hot water tank was full, a few things could happen.
Case 1: If your pump was very powerful, and nobody in the apartments was drawing any hot or cold water.
- You could push the heated water through the water heater and through the cold water intake pipe, back towards the water company (until it got used by the nearest house in the neighborhood)
- Before doing this, the water might come out someplace else (ie the toilet, as was suggested)
Luckily this design would ALWAYS draw cold water when it is drawing hot water, so this case would never happen. But it is demonstrative nonetheless.
Case 2: If your pump was moderately powerful, and nobody was drawing any hot or cold water EXCEPT the cold water being drawn by the AC.
- The hot fins of the AC are "cooled" by a loop of fairly hot water. The size of the loop is the distance from the hot water pipe in the apartment to the hot water heater, plus the distance from the cold water intake for the heater to the cold water pipe in the apartment.
Case 3: Hot water is being drawn by somebody who is connected to the hot water heater.
- Almost-free hot water!
Case 4: Cold water is being drawn by somebody in the immediate area
- It might be fairly warm, but at least it might take some heat out of the "loop" which is otherwise formed.
Nothing explodes, but the AC might overheat and cycle off. Its efficiency would go down. And since the water isn't much cooler than the fins, the hot coils would probably start heating up the indoor air, depending on how well you insulated them.
Agreed.
I'll be honest, I don't have the inclination. Not because it's not a good idea, but because it's not suited for an apartment. In a home, I think it would be stupid not to use this idea in at least some form (see the "counterflow heat exchanger" also). I'm not going to be arrogant and claim that I do have all the required knowledge, but I don't see it being that difficult. Someone should try it.
I think for now, I'm just going to buy a water-cooled unit.
BTW: Some "Split systems" have user-removable refrigerant lines between the compressor/condenser and the evaporator. Nonetheless, I'd rather not have rigid freon lines running through a house unless they were behind the walls.
Taking water out of the cold water line and putting it back in would be Ok only if it does not mix with anything else. If it goes into an open system, then is put back into the water lines, there is possible contamination of the potable water system. To do what you want to do probably requires a vented double wall heat exchanger to prevent contamination. This reduces efficiency but protects our drinking drinking water. You may have to install a back flow preventer also, which will make pumping water back into the water lines impossible. Better consult your code authority.
Stretch
PS: If you wan to use a chilled water system, it would be cheaper to buy one than make one from scratch. Or use a water source heat pump.
It doesn't. The cold water would flow through a sealed copper pipe surrounded by unpressurized propylene glycol in a heat exchanger and back into the hot supply pipe via a small circulation pump.
For instance, if you use 50K Btu/h of AC, you put 0.2gpmx60minx5h = 120 gallons of hot water into the system. If you and your 11 neighbors use
12x50K = 600K Btu/day of hot water, you remove about 600K/(110-60)/8 =
1500 gallons/day from the system, at different times.
With a very large hot water tank, the cold water would never get warm. As the tank size shrinks, this would happen more often. We could do simulations with various tank sizes and random hot water bursts to compare the value of the energy savings with the cost of the occasional hot water dumped. Or just try it. If the neighbors are taking showers or washing dishes vs making ice cubes, they might enjoy warmer cold water and lower bills.
Sure, in the unlikely event that there's only 0.2 gallons of cold water in the bottom of the water heater and nobody's using hot water. I doubt the T&P valve would care.
The dump could happen automatically, with a thermostat on the incoming cold water line to the OP's apartment and a solenoid valve to let some hot water flow into the drain. The net benefits would be positive, IMO.
I might opt for the 400' of pipe in the kiddy pool, in series with the cold input to the water heater in summertime, and no extra water consumption.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.