Towards a better tinyhouse

Inventing to freedom?

<$160 year round solar heater for a 250 sq ft tinyhouse

with 3 comments

Okay, remember that gelified water solar collector idea? I decided to take some time to put together a spreadsheet and check out the details. Which took way longer than it should have, but…

Anyway it turns out there is not much point in gellified water, due to the high cost of the gelling agents. Polyvinyl alcohol is $21 per Kg or so apparently, and I checked agar which is triple that, but if you could get the cost of the gelled water to below about $60 per ton it might be cheaper for this application. If you needed higher temperatures then it might make sense. Bizarrely I found a patent that describes the idea here, and you won’t believe me but I found it while looking for the thermal conductance of gelled water. Notice that nowhere does it actually have that information in it – search engines can be so weird. And annoying.

Anyway, so this uses greenhouse plastic. The spreadsheet lets you easily customize it to your own climate, but I chose to use the data for a challenging climate, Minneapolis, Minnesota. This design described below keeps a 250 sq foot, with loft, just built to code, with a little ventilation 0.3 changes per hour, but no heat exchanger, tinyhouse warm all winter in a typical winter. This is using the typical meteorological year data, so it might be a good idea to double check with a more severe winter before you built. A link to a place you can get the data is below, but it costs $20 or so, and I didn’t want to pay that.

If you just want to get your results for the typical meteorological year, you can just cut and paste the information into the spreadsheet after downloading it from the the site linked to below.

I explain how it works below, but the spreadsheet is pretty accurate and factors in solar radiation data, ambient temperature, ventilation rate, the insulation values of the walls and windows, and then you simply enter values for the solar collector and it will return information about system performance, and draw you a graph like the one below. Adjust the parameters, like the size of the collector etc. and you’re all set.

The way the system, and the spreadsheet works is this:

There is a shallow insulated box on the ground. You can set the insulation value. In the specific system I came up with here it is a U value of 0.4 which is r-2.5 so a board of polystyrene foam. the bottom is just foam and rest right on the ground. The sides are probably about 15 cm high and made with 6 by 1 wood boards.

Okay, now there is a plastic liner of greenhouse plastic so this box can now hold water. The box covers 10 square meters. Probably 2 meters wide by 5 meters long. There is some water in the box, a layer a centimeter thick or so doesn’t really matter, and some tubes so the water can flow in an out of the box through them, in along one side and out along the other. The water has antifreeze of some sort in it, probably something harmless and convenient like urea or sugar and some salt or sodium benzoate or something (used to preserve pop) to prevent microbes from growing. This is the water that gets heated up and used to transfer heat around. It could be dyed black, or you could just put some black construction paper on or paint the very bottom of the inside of the box, in contact with this water.

The tubes do not go through the sides of the box, they just go over the edge, so no seals are needed or anything.

Next, directly above this is the insulating transparent layer. You can set it’s light transmission and U value in the spreadsheet. It is set to a u of 4.4 and 0.9 transmission or so right now, which is fine for 2 layers of greenhouse plastic (which is more like u 4). Actually you change the “thickness of gel” cell to change the u value, you might want to change that.

Okay, that’s the actual collector thing. There is also a reflector, and this really helps to reduce the size of the collector and increase output in an economical way. This just something flat with mylar attached. You could just build a frame out of lumber and put the mylar over the frame, but it might flap in the wind. There must be a better reflective material available that is rigid enough to not flap. Otherwise just take some 1/8 or 1/4 inch ply whichever is cheapest and glue the mylar or Al foil or whatever reflective stuff there is to that. The fact that not all the light is reflected from the reflector is accounted for in the spreadsheet and you can set the reflectance.

Also, you’ll notice the reflector is at a certain angle. This is accounted for in the spreadsheet, and you can adjust the angle. You can see it helps to give you a bit more power, and you would probably a little roof extending over the top with a bit more plywood to keep the snow off of the collector, which I forgot to add to the pic, although it will block just a bit of the diffuse light from the sky, it also adds some more direct sunlight. The fact that the reflector also blocks some of the sky is also accounted for, and it still gives you a big net boost.

The reflector is bigger than the actual collector part, probably extends 1.5 meters in either direction to give you a bit more boost in the morning and evening, and a meter longer than the actual collector. It is assumed in the spreadsheet that the spot of light is bigger than the actual collector, so when it moves due to sun movement it is still covering the actual collector. This is not accurate during the early morning or evening but you can see looking at the solar data that almost no sunlight is available then anyway, so this fact doesn’t cause any serious inaccuracies.

Okay, then you have an insulated storage tank, probably just some 2 by 4s, 1/4 inch plywood and lined with some polystyrene foam and some Great Stuff PU foam sealant to seal around the PE foam, and some more greenhouse plastic to line the box with. It is a cube 2 meters by 2 meters by 2 meters. So lots of water, but water’s cheap.

Next, there is some thin walled plastic or metal tubing on the the side and towards the bottom of the inside of the box immersed in the water tank. The heated water flows through this and then back to the collector, by convection.

This is just to allow the water heater (the solar collector) to heat the water in the tank, without just running the water in the tank through the heater. If we just ran the water though the heater we would have to do something about the water pressure that results from the height difference of the water, and also the whole 8 cubic meters of water in the storage tank would have to have antifreeze stuff added to it which would be too expensive and probably kind of messy or damaging to the soil in case of leaks. So there is storage water, and there is transfer water (well transfer fluid since it is more than just water, since it has the antifreeze).

Then, the water from the storage tank is capable of flowing through a radiator inside the house, probably with some sort of simple thermostat control thing of some sort, you’d have to figure that out. The radiator could be a large plastic garbage bag or two coiled up a bit and with some spacers to allow some air to flow between the layers of the coil. It should to have a surface area of more than 2 meters or so. A radiator from a car or from one of those old hot water heat systems might work, just check the surface area. It needs to be able to transfer more than 600 watts with a temperature differential of 20 degrees, which is not that hard. You might be able to make one with, I was thinking, a piece of aluminum flex duct, just flatten it and bend it in a u shape, with the top of the U above the top of the water level in the storage tank, so there are not really any watertight seals to make on it. As you can see there are many options for a cheap radiator if you give it some thought anyway.

Water flows through the radiator again by convection. I still have to check the physics to see what size of and length tubing would be required/okay for adequate convection but we’re talking ten milliliters per second here so it’s not much and shouldn’t be a problem. For the collector side it is hundred and something mls per second peak but I would think not a problem at all with adequate size and short tubing.

Also, when water is heated in the tank by the heating tubing coil it rises to the top of the tank instead of being all evenly mixed and stirred, so it should be more or less a laminar flow tank, like a hot water tank works. This just makes the radiator a bit smaller because the water going in is hotter so the average temp across the surface of the rad is higher. Basically if the whole tank was mixed the energy would still be useable (in this particular scenario) but held as a lot of warm water rather than a smaller amount of hot water. But it doesn’t really make that much difference.

Probably for tubes that enter and exit from the water storage tank, it would be better to have fittings and put them right through the side wall, but you might be able to get away with putting them up and over the side wall thus avoiding the fittings and associated potential leaks and extra construction work.

The plastic would probably cost $60 at most, probably less I think I remember seeing it for $3 a square meter somewhere, then 2 layers. Plywood lumber and tubing and foam etc I would think less than $100.

And that’s it! The system needs no electricity, is dirt cheap to make compared to alternatives, takes no fuel to carry or buy, you could use the water tank to store drinking or cooking water if you make things out of suitable materials and in a typical year it works without leaving a single day without heat! Not bad. Again I would caution that this is for typical not an extreme year. It’s also in a very cold and not very sunny climate, Minneapolis minnesota, too, though, so clearly this sort of system is totally practical.

In the summer you would fold the reflector thing down, over top of the collector, so that the plastic is not exposed to sunlight unnecessarily. Judging from the amount of sunlight exposure this would save the plastic from (shown by the red line in the graph), and that this plastic should last 3-4 years if it were fully exposed year round (I think that’s what they said) this should easily greatly extend the life of the plastic. Another thing that could make it last for more than 10 years, eyeballing it, would be to split the collector into more than one unit, each of which can separately be folded in this way. As you can see from the graph, you only need all 10 meters during the dead of winter, so by folding the ones you don’t need when you don’t need them you could eliminate a lot of sun exposure from them.

One more thing I would like to add if I had time is better simulation of the water tank heat storage, with the capacity to take into account that the tank can store extra heat by increasing the assumed storage temperature during very sunny periods, and that it’s heat can be negative if the water gets colder than the indoor temperature because then energy has to be added to heat it up before it can be used to store useable heat.

Also, this started out as a design that was going to use gelled water, which wa originally why the collector was on the ground (water heavy), but then I switched to greenhouse plastic after finding out the cost of gelling agents, and I think putting the collector on the ground os a pretty good place, you get the protection from snow so it’ll keep working rain or shine, and you can fold it over during summer. You might be able to reduce the cost a bit though with more reflector and less collector but… also you can use glass or whatever if you want, just put the right U value and transmission value into the spreadsheet.

Here’s a quick sketch of it:

A graph, you can see there is plenty of heat in the summer if you wanted to use it for heating hot water for showers etc. or a heat powered air conditioner:

spreadsheet is here :

Polyvinyl alcohol, 20 lbs for $211:

example double glazed window:

free climate data for typical year just cut and paste:

climate data for a variety of years but not free: and


Written by gregor

January 22, 2011 at 09:31

Posted in Uncategorized

3 Responses

Subscribe to comments with RSS.

  1. This is very interesting. I wonder if it could be applied to in floor radiant heat? Otherwise, I’m sure a radiator type heat would work well. Do you live in Minneapolis? If you do, I’m just curious if you could help me put one of these together to see if it works.


    January 23, 2011 at 14:44

    • I don’t live in Mineapolis, actually, I live on Ottawa, but I chose Mineapolis because it is a challenging climate, and it is similar to Ottawa’s.

      But I’d be really interested to see someone do this, it’s why I put the effort into making this rough design, so I would be happy to help with the design of it if you need something, though I can’t help with construction it should be relatively easy to build.


      January 24, 2011 at 08:25

  2. Good Info, simak terus di <$160 year round solar heater for a 250 sq ft tinyhouse | Towards a better tinyhouse

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: