inexpensive off grid air conditioning for a tinyhouse
Okay, so I was wondering how to do this, and I put together a spreadsheet. I’m no spreadsheet ninja, and this is the first time I have used Calc, but you can put your own data in, and see for yourself how these things could apply to your own situation. I got my data from here . It would be interesting to look at a hotter climate, and see what sort of approach would work there, and I tried to find some data for California, but no dice. The data I use here is for July, the hottest month here. See the last graph for the actual temp. data plotted.
Scroll way to the right, there are the graphs and stuff, and the assumptions like the U value of the walls and stuff, that you can adjust. I’m assuming a third of the walls are getting pretty hot from the sun, 45 degrees, but it turns out it doesn’t have that much effect anyway.
There are a couple of ways of cooling without electricity, and you have no doubt heard of thermal mass storage, which I have been calling thermal closeting, though turns out that is not a very common term for it. Whoops. Also, evaporative cooling . The spreadsheet includes a basic simulation of both these approaches. Definitely basic. But it tells you more or less what you can expect. But first we need to know the cooling requirements:
The reason the graphs are so spiky, rather than having a smoother increase in the amount of cooling required as the ambient temperature goes up, is that it turns out, surprisingly, that by far the biggest load is the heat produced inside the house by electronics, cooking, all that stuff, plus solar gain from the windows. By far. I assumed it is 200W at night and 400w during the day for electronics, and a sort of arbitrary 100w for solar. If the temperature is below 23 degrees out, hypothetically, you can just dissipate the heat by dumping it outdoors. When the temperature is higher than that (or whatever you consider to be the maximum), all of a sudden, you need to produce that lower temperature, to remove the heat. The contribution to the cooling requirements that dehumidification and the need for ventilation imposes do change more smoothly.
The interior of the house isn’t enough thermal mass to take full advantage of thermal mass storage, as you can see below. This graph is with a 1 ton of water, I would think more than the interior of a tinyhouse, and it still leaves some days unconditioned (the heat capacity of wood is about half of water’s). Even if you are willing to settle for 25 degrees, it doesn’t get you though day 5 or 6, because the temperature the previous night was too high. Plus, if you are using the interior of the house, that ignores the fact that you are cooling the thermal mass to an uncomfortably low temperature sometimes, and need to blow a lot of air through the house to recharge the mass. Also, I did not include anything to take into account the possibility of it helping with the latent heat needs, because it would have no effect anyway in this climate. It would have to go below about 14 degrees during the previous night. Maybe in the desert, though.
Strictly speaking, there should be some factoring in of the thermal resistance between the thermal reservoir and the house, but I ignored that for now. In effect, that would slow the rate at which the reservoir absorbs heat, a couple degrees before it was full. You could use phase change material, and that could make it a bit more practical than a ton of water, but to make a good decision about what types and how much you’d have to look at more data, I think, but 20 kg of phase change material would really help with those times when the night temperature is really close to the max. permissible indoor temperature.
Oh, by the way, I got some bad news about the heat exchangers. Turns out the manufacturer’s lie through their teeth about the efficiencies http://www.hvi.org/assets/pdfs/CPD/HVICPD_Sec3_Sept2010.pdf. Looks like 60% for sensible heat, and 40% for water vapor is more reasonable. It does go up, though, if the airflow is lower, and in this calculation the airflow is 8 CFM (223 LPM), rather than the 120 or so that the exchangers are usually tested at, so you can change the heat exchanger efficiency in the sheet if you want.
One thing is, though, that it might turn out that they are pretty expensive, though. In that case you would go for the smallest one around, I guess.
Another, kind of unfortunate effect of the fact that it is the heat produced inside that is the main problem, is that you can’t easily distribute the cooling power through the ventilation system as is done in a forced air system. The 8 CFM of air would have to be pretty cold to transfer the necessary amount of energy, especially during the daytime. Which would pretty much require electrical cooling.
Instead you would have to either put cooling coils in the wall or something, or, I think probably more practical, is to use a fan blowing over an air-water heat exchanger. I remember seeing a tinyhouse with a ceiling fan once. That could work fine, just put the exchanger above it. Or, I noticed there are fairly cheap “evaporative air conditioners” out there, which by all accounts don’t work for jack, but they have the air-water heat exchanger and a fan in them. Maybe that or some other consumer product could be used.
They might work as an evaporative cooler, too, but you have to put the evaporator outside…. The problem with their design is that the evaporator is indoors, so it increases the humidity in exchange for reducing the sensible heat http://en.wikipedia.org/wiki/Sensible_heat (eg decreasing the temperature of the air), which is exactly the opposite of what you want.
It’s pretty amazing how much cooling power you can get from water, though, the system below uses about 10-15 liters per hot day. I plotted it in 100s of ml so it would show up on the graph. You can see it’s far from perfect, though, it’s no good on the days humidity is high. This device is made to cool the water, not the air, then you pipe the water around wherever and use it to absorb heat. Unlike a swamp cooler, it does not add water to the indoor air.
The system in the spreadsheet, uses a regenerator to improve it’s efficiency. To see what would happen without it, set the efficiency to zero. Basically something like the Ultra cheap ERV I just posted, but you don’t have to worry about the smell of the regenerator media, so you could use polyester quilt filler material, or sewn together sweatshirts or something. But it can’t ever cool below the outdoor dew point – or rather an outdoor dew point.
An idea occurred to me, that if you did a whole bunch of evaporative cooling during the night, when the dew point is lower, using it to cool down a thermal reservoir to the ambient dew point during the night, then it could be cold enough to be used to do some dehumidification during the day (when the dew point rises). Unfortunately it looks like that wouldn’t do much good in this climate. What you really want to do is to remove water from the indoor air, as it comes out of the heat exchanger (heading to the interior of the house). The dew point during the previous night has to go below the dew point of the indoor air the next day before it even starts working. Well, I guess you could dehumidify the air going into the exchanger from the outside, and that could help just a bit on some days. Hm, sounds like worth looking into, if it helps enough to open up other options elsewhere….
Another, more useful idea (for me) is to use an electrical heat pump like a compressor cooler to handle the dehumidification. If you are off grid, chances are you will have a solar panel somewhere, which means you’ll have extra power available in the summer. Might as well put it to use. The compressor in a cooler might not be powerful enough, though, but it could be worth checking. You’d need to have a way to get it to do just the dehumidifying, without the need to release the cold air (since it has to cool the air down first before it starts to extract water). You could use a… I know, I know, not another one, but, a counterflow heat exchanger. The air goes through it, into the cold chamber where water condenses out, then back through the exchanger again, and out. This one would be a cheap one, because you only need it to exchange sensible heat. You can make such an exhanger with Al tape and Al foil and some cardboard, just fold it like an accordion, and one fluid flow goes on one side, and the other on the other side.
Heck, if you had a big enough solar collector, after thermal mass storage, you might be able to use an electrical air conditioner, I remember seeing a 5000 BTU one somewhere, and you can get really small 12v ones for truck cabs, as a low cost replacement for a broken main AC system. 4000 BTU/hr is about 1000 watts, IIRC, so that’s just about right, and it should take 300 watts or so to run.
Another way would be to use a solar powered desiccant system . Liquid, or the desiccant wheel thing if you can swing it, they don’t seem to say so on that web site but with heat input, the desiccant wheel can be used to actually pump water vapor, too. You heat the air that is on it’s way out of the system directly before it passes through the wheel. Apparently they have an efficiency of 10% or something, but you don’t need much power here anyway. With the wheel system you can use a lower temperature, so you can use a flat panel collector, whereas I think the liquid ones need vacuum tube collectors. Might be hard to get a hold of a suitable one, or the carbon stuff, though. It doesn’t have to be a wheel, you can use 2 or 4 stationary beds of adsorbent material, and use an air switch to switch the direction of airflow between them. I wonder if you could make you own adsorbent by dousing some activated carbon with calcium chloride.
For the liquid system, take some tangled fish netting, or something similar, and by dripping water on it, that can form an evaporator (or absorber), just as a way to present a large surface area of water to the air. Then the water goes off to be heated by the sun, driving the water out, and back to absorb some more.
The last cool possibility that I was thinking of is a personal cooling device of some sort. As long as you can keep your body temperature down, you stay comfortable. I don’t know if you’ve ever walked out of an over conditioned theater into a really hot day, but it’s actually pretty pleasant at first. Until your body temperature rises because you can’t dissipate enough heat (or, if it is above 37 degrees, because heat is even being conducted into you, too).
“Personal cooling system” finds plenty of examples. I wasn’t impressed with any of the ones I saw, though. Not very efficient, so you have to lug around a big cooling reservoir. Also, I wonder if having the cold tubing against your skin would be very comfortable.
It seems like the best way from a comfort standpoint would be to only have air in contact with the skin. Basically, to shrink the air conditioned room down to the size of a vest. It would have comfortable elastic cuffs around the arms and waist, and a turtleneck around the neck to seal it. There is no need for any ventilation, though, because you are not breathing the air. You can just have a fan or something blow air around, through a tube on the back of the garmet to the right side, then into the vest, across your torso, gets sucked in the other side, to the cooling coils or whatever, then back for another run. Or maybe use tubing made of a water vapor permeable material (the tubing does not contact your skin), and flow cold water or a desiccant material through them. Water vapor then diffuses into the tubes, and the heat, and flows away, keeping everything comfy. It can then be cooled with evaporation if a desiccant, or ice for water – the advantage of using a desiccant is that it does not have to be at a low temperature to absorb water, so it can be cooled purely by evaporation. Evaporation gets you about 10 times as much cooling power for the same weight, and you don’t need a fridge to keep producing ice. Mind you, if you are indoors, that would raise the humidity…
Lastly, personally, I noticed something this summer, which struck me as pretty handy and useful – I, for reasons I won’t go into, had a bunch of wet clothes laid out on my bed. It was really uncomfortably hot and humid. The air conditioning is intermittent here, and it was not turned on, though I don’t remember the exact temperatures, checking the weather records wouldn’t have meant much, because I’m pretty sure it was hotter inside than outside (I didn’t want to open the windows because the noise is really bad here). I was resigned to spend a night sweating. So, anyway, I went to go to bed, and they were still wet, and, being pretty tired, I was faced with a dilemma: Put clothes away or no? At first I assumed I better put them away, because I fuzzily recalled sleeping on wet clothes is uncomfortable (yes, this had happened before). But then I realized that was only because it was so cold. So I tried it, and it was actually really comfortable all night. I slept better than usual, even. It’s amazing how they seem to stay wet and cold all night long (previously very annoying). I think this could be really useful in Ottawa here, where there are usually only a relatively few nights a year where it is really too hot, or for tinyhouses with no AC at all.
They weren’t fully wet, they had been in the dryer, but the “automatic dry” feature screwed up, and they were not dried fully. They were definitely heavier, just to give you an idea of the water content. Wetter clothes would work fine, I guess, just from after the spin cycle. You would think mildew would grow, but it didn’t. Having just been washed inhibited it enough, I think, and maybe because there was still some detergent on them.
Essentially this is a bit like a wet bulb thermometer. You are the thermometer, and of course the clothes are the wet cloth.