How to calculate air conditioning efficiency?

August 16, 2007 6:20 PM

Need some advice on how to calculate 'efficiency' of a potential home-brew air conditioning set up.

My workroom - well-insulated, no air conditioning - is at 30 degrees (86F) this morning, and this will climb to around 32+ this afternoon.

Just outside, directly below my windows (I'm on a very steep slope) and seven meters below my floor level, is a small river. It's not so deep, has good flow though, but with the hot sun on it all day it's not so cold. I just checked, and it's 23 degrees (73F). That'll probably climb a couple of degrees later in the day.

Is it worth my while to investigate the possibility of working up some kind of system where I pump water up from the river, and run it through a heat exchanger to cool this room? Or am I going to spend so much electricity on a pump and the exchanger, to negate the benefit I will get from that 7~8 degree difference in temperature? (In other words, is there no advantage to just buying/using a commercial air conditioner?)

posted by woodblock100 to home & garden (12 comments total)

The principal benefit of a conventional air conditioner that your proposed alternative will not achieve is de-humidification. The temperature difference you'd be working with from the river water simply wouldn't be great enough to get significant dehumidification effect, and since drier air feels substantially cooler, a room cooled down to the temperature of your river water (assuming you could pump so much river water, and had such an efficient heat exchanger that such a theoretically perfect cooling state could be achieved) would still not feel as cool as a room air-conditioned and dehumidified to even a few degrees warmer than that, by a conventional air conditioner.

So, buy a conventional air conditioner, to cool the room air.

However, in some climates, you can benefit by using water in a roof mister system, substantially reducing the heat developed and conducted through to attics, by a roof exposed to direct sunlight. I've seen fairly effective low tech versions of this made of nothing more than a couple of soaker hoses laid just below roof ridge lines, and held in place by a galvanized roofing nail every 5 or so feet. It only takes a few dozen gallons of water a day to get rid of a lot of roof heat, when the water is entirely evaporated to the surrounding atmosphere, as a working fluid.

The reason water is much more effective when used as a cooling fluid in this way, than it would be in your proposed setup, is that the temperature differential that is creating phase change in the water (as working fluid) is many, many times greater on a roof mister than it is in your non-phase change proposed cooler. So, it might make economic sense to pump some water up to your roof, to cool the roof by evaporation, thus lessening the amount of conducted heat your air conditioner would have to deal with substantially.
posted by paulsc at 6:50 PM on August 16, 2007

The temperature difference you'd be working with from the river water simply wouldn't be great enough

Yes, that's what I was afraid of ... A second option I was considering was forgetting about pumping (heavy) water up from the river, but just dropping some kind of duct down there that would scoop up (light) air from just above the water surface. But the dampness is kind of scaring me off that kind of attempted solution ...

you can benefit by using water in a roof mister system

Hadn't thought that far, although what we have done on hot days is drop buckets down into the river, haul them up to street level, and splash over the street in front of the house. With a few of the neighbours doing the same thing, the temperature out there drops quite a bit. But as far as pumping water all the way up to the roof, I dunno ... Here's a quick snapshot of the building, taken from the other side of the river. It's a long way up to the roof ...
posted by woodblock100 at 7:23 PM on August 16, 2007

Depending on where you're located you will need a water right to legally pump from the river.
posted by fshgrl at 8:54 PM on August 16, 2007

In North America, a BTU is the energy required to raise one pound of water one degree Farenheit. A gallon of water weighs about 8.4 pounds, and you're working with water having a temperature difference of something like 10 degrees Farenheit. So, for your situation, each gallon of water has an effective BTU absorption rate of something like 10 degrees X 8.4 pounds or 84 BTU. If you want to take out the amount of heat a small room air-conditioner would, say 12,500 BTU/hour, you'd need to pump a minimum of 148.8 gallons/hour of river water through your heat exchangers, assuming your system was thermally perfect. Let's say you'd like to be able to pump 200 gph (to have some excess capacity, because your thermal efficiency could never be 100%), against a head of approximately 25 feet, you'd be able to do that with a 1/2 HP shallow well pump. With a high pressure model pump, you could actually get 5.0 gpm at 50 psi outlet pressure, so assuming you put a pressure tank at the top of the pump lift, you'd have enough pump capacity to run the pump at about a 66% duty cycle (which is pretty heavy use), for the flow rate you'd be trying to achieve. Your power cost for moving that much water with such a system would be a little greater than for operating a 12,500 BTU air conditioner, assuming you used something like 3/4" diameter piping, because the water pump has to overcome greater frictional losses in moving the water, than the air-conditioner has to deal with in moving phase change refrigerant, with essentially zero pressure head. And the water system provides no de-humidification, as previously noted.

That's the problem with cooling systems where no phase change occurs in the working fluid - they're just not very efficient, thermally. Whereas, as the second link of my previous post showed, 1 and 1/2 gallons of water going through evaporative phase change to 100 degree C, can carry off 8,500 BTUs. So, if you could go with 1/2 hp deep well pump, and pump only 2 gallons per hour to your roof for total evaporation, you'd get about the same cooling effect from 1% of the amount of water, because of phase change dynamics. Your energy cost for pumping such a small amount of water an additional 25 feet vertically (3rd floor roof), would be a small fraction of that for circulating 200 gph to half that height, and returning it to the river.

The only way of materially improving the efficiency of the 200 gph water pumping system, would be to recover some of the head energy in the water column, as the warmed water was returned to the river. I suppose you could make some kind of clutched recovery turbine, mechanically ganged to your well pump to do this, but it is not a normal kit of hardware I could recommend as off the shelf parts. And, there are going to be conditions of vapor lock in such a system, that are going to require check valves or other measures to keep the downline from going dry, and thus losing its prime.

The power costs for achieving similar cooling rates by moving air as the working fluid, as your reply contemplates, would be several times the energy requirements of even a water circulating system, owing to the low density and poor thermal transfer characteristics of air as a working fluid. Figure several horsepower in blower capacity, at a minimum, if your external duct work can be well insulated.
posted by paulsc at 9:43 PM on August 16, 2007 [1 favorite]

15F is a very low grade thermal difference. To put it in perspective high efficiency air to air heat exchangers used in intercoolers for turbos are hard pressed to cool the intake charge to within 20-30F degrees of ambient. And that's with ambient air whipping by at 50 mph. You'd have to have a massive radiator to get any reasonable temperature change in your room. If you used a heat exchanger at both ends you wouldn't need a great deal of electricity to power your pump because in a closed circuit you only have to over come friction losses and not an eight metre head. The larger the pipes (within reason) the better in that case. Your water-water exchanger will be much smaller than your water-air exchanger because water is a much better exchange medium; air is practically an insulator.

Alternatively because you have good flow if you wanted you could use the flow to pump the water. There are several microhead pumps or generators available. They can be as simple as a DC trolling motor mounted in the stream and wired to your recirculating pump. A hydraulic ram pump would pump the water directly. Viola, free water pumping if you discount the capital costs. Do a search on micro hydro.

Your water is a great source for a heat sink for a water cooled condensing unit which is what I would do with it.
posted by Mitheral at 10:03 PM on August 16, 2007

That's the problem with cooling systems where no phase change occurs in the working fluid - they're just not very efficient, thermally.

This indeed seems like the key point here; without a level of sophistication/efficiency in the system (that I would be unable to build for myself) it just doesn't work out to be viable.

As for the roof misting concept, it does sound like something that might be reachable, at a stretch. Problem for me, re the current situation of 'trying to cool my hot workshop', is that what happens way up there under the roof - four levels and 12 meters above me - has no relevance at all to the workshop situation. It's quite a bizarre structure (here's a sketch of it). The house up top has no effective insulation at all, and my renovation have started with the bottom basement, and - at current rates of progress - will take many years.

Anyway, thanks very much for the time spent on the calculations Paul; you've saved me from wasting bunch of time/money on this!

you will need a water right ...

I have no rights at all of course (this is in Tokyo). All I was hoping was to work out a 'solution' that would involve dropping an intake hose down there for a few hours each day during hot afternoons, and then hauling it back out of the way when the system wasn't in use ... (Not so much different from what we do now with the buckets ...)
posted by woodblock100 at 10:17 PM on August 16, 2007

I have nothing to contribute so should probably keep my mouth shut, but man, that's one lovely house, woodblock100! And after reading the details on your site, what an incredible deal! So much cheaper than this rabbit hole my family and I live in now... I am jealous indeed.
posted by misozaki at 12:17 AM on August 17, 2007

Ah, after further reading, I now see that a LOT of manual labor has gone into making your home what it is now! Amazing.
posted by misozaki at 12:30 AM on August 17, 2007

man, that's one lovely house, woodblock100!

Thanks Misozaki ... long time no see! :-)

But I tell you ... since posting this question in the morning, the temperature has been climbing steadily, and the thermometer on my workbench now reads 33 ... And yet as I sit here working, just outside the window I see that stream rippling by. Talk about frustrating! But no point arguing with the laws of physics ...
posted by woodblock100 at 12:41 AM on August 17, 2007

Would appear to me that the room has a window? Is that an east/west window? Put an awning on that puppy! Keep Mr. Sunshine OUT.

Also, you're getting stack effect that high up, your "well insulated" room is keeping the hot air inside. Install a gable fan or 2 and get that poop OUT. Drawing cooler up up through the house is your best bet AND should help dehumidify since you'll be moving the air.

Also, consider putting some radiant barrier up there to prevent thermal gain through the roof. Radiant barrier CAN be installed INSIDE an attic space, shiny side facing OUT, and best installed with a layer of plastic on the room-side behind it to prevent dust buildup on the reflecting surface.
posted by TomMelee at 6:50 AM on August 17, 2007

Put an awning on that puppy! Keep Mr. Sunshine OUT.

Room is a printmaking workshop ... faces north. Solar gain is not a problem.

Drawing cooler up up through the house is your best bet

I am indeed trying to do this. At these times of day, the two casement windows are wide open, and there is usually good airflow up the stairwell and out at a higher level. That isn't doing much for dehumidification though, because the air that comes in is coming from the river 'valley', and is pretty saturated.

consider putting some radiant barrier up there to prevent thermal gain through the roof

When my renovations get that far up, the attic will indeed need to be insulated, as the top floor of the building is currently pretty much uninhabitable in summer. It's going to be interesting to try and figure out a way to use all that roof sunshine in winter, to bring as much energy into the building as possible, yet still keep cool in summer. But that's all for the future ...
posted by woodblock100 at 2:19 PM on August 17, 2007

paulsc has a good suggestion with the soaker hose. We did this setup to make our tin can trailer home livable in the humid sunny southeast. Didn't lower the humidity, but sure helped with the radiant heating.

We set the sprinkler hose (the kind with little holes punched in it) on the ridge. Set it with holes facing up - we had a small amount of water to dribble off the eves and it worked well enough to make it through the summers. We adjusted it so that the water off the eve was just a bit warm to the touch. (our themostat, LOL)That way we weren't wasting too much water/energy. Course, it helped our plantings at the edge of the abode do quite well with the increased/consistent water.
posted by mightshould at 5:07 AM on August 18, 2007