So, I’ve been wondering for a while about whether you can mostly heat your house with a (large) solar hot water system, plus a good sized storage tank to bridge you through cold periods, plus a hydronic (under floor or radiator) system.
I’ve been researching some more today, and probably am not a lot closer to knowing. But what I have learned today I’ll include here, over time I’m going to post some more on this as I form more coherent ideas.
Let’s start off with the type of solar hot water. There are broadly two types: evacuated tube and flat plate.
A flat plate is largely a board with pipes of water on it, and a flat plate of glass over the top. Often they have insulation, special coatings to reduce radiation, and high performance piping etc. But at the bottom they’re a box painted black with black pipes in it. These are quite cost effective, and they generate quite a lot of heat per sq m of roof in summer. Their performance tends to decrease when the outside air is cold (winter) as the water and the box itself are not well insulated from the outside air. Logically this could be somewhat improved with double glazing, although I haven’t seen any units like that.
An evacuated tube unit is essentially a series of thermos flasks. Each tube is double walled glass, with a vacuum between them. The inside glass is typically black in colour to capture heat, the outside glass is clear. The inside of the tube has some sort of fluid in it – water or something that creates less pressure when it boils. The inside tube gets hot from the sun, the liquid gets hot and rises to the top. The very top of the tube is encased in copper, and slides into a manifold full of cold water. So the fluid gets hot, rises to the top, gets cooled by the water, runs back to the bottom, and so on. Upshot is the heat is transferred to the water in the manifold. Because the manifold is insulated, and the evacuated tubes themselves are like a thermos, they are less impacted by outside temperature, and they generate some level of heat even on an overcast day. Conversely, because they have a black inside tube and a clear outside tube, only a percentage of their area is drawing in heat. So they tend to require a larger roof area for a given output.
The decision between the two currently looks to be a religious one. There are studies in Europe showing that evacuated tubes perform better in the cold, except for when they get snowed on. Because snow slides of flat plates easier than off tubes, once it snows the tubes will tend to stay covered. Of course, it doesn’t snow in Australia, so that’s a bit of a pointless argument over here. In general it appears that an evacuated tube setup will generate more heat in adverse conditions than a flat plate. And since my plan is to use it for heating, and you need heat more on adverse days than on clear and sunny days, I think evacuated tubes are the go.
Next, the question of how much heat you can get. According to wikipedia, most of Australia gets around 4kwh per sqm per day in the winter months. A more detailed calculation for Canberra can be derived here, this site provides a bunch of photovoltaic calculations, but also provides the raw insolation numbers. The minimum monthly insolation is around 3.8kwh per sqm per day in mid-winter, this value is reached with the collectors angled from 50 to 65 degrees*. The calculations suggest that the evacuated tubes are probably around 50% efficient, so I’d be getting 2kwh per sqm per day in winter.
In terms of heating our house, running the old standard “1 bar heater” gives you 1kw of heat. So 2kwh would be as much heat as a one bar heater running for 2 hours. Or, in other words, not much. I reckon a very well insulated house could be heated with a 1 bar heater running continuously, or 24kwh per day. A house more like our one probably needs 2×1 bar heaters, or more, so around 48kwh per day. Cross checking this with our electricity bill, looks like we use 30kwh per day in summer, 90kwh per day in winter, a 60kwh difference. But that’s using a heat pump, which gives about 2kw of heat for every 1kw of power input. So the range is probably between 50kwh per day and 100kwh per day. This means you’d need 24 to 50 sqm of collector to sensibly heat your house. It’s not that that’s a problem, we have 200sqm of roof, and probably 50sqm of roof that could have collectors on it. But there’s quite a bit of money in getting that much heat collection.
A 30 tube collector @ 2m long and effective collection area of 50mm per tube gives 6 sqm of collection, and costs around $1,000. So there’d be $4,000 of tubes in it. There’s probably another $3-4,000 in other equipment including the storage tank, pumps etc, and there’s also some power usage from the pumps. My gut feel is that heating our house with purely solar, or even mostly solar, is probably not economic on these calculations. Having said that, we’re spending about $1,000 for the coldest 3 months of the year on the power bill, call it the same again spread over the remainder of the year, if we could halve that there’d be $1,000 a year in savings.
*On the question of panel angle, the physics here is that the most collection happens when the tubes are directly perpendicular to the sun, maximising their collection. The problem is that the sun keeps moving around, so its lower in the morning, high in the middle of the day, lower in the afternoon. Noting that the overall insolation remains the same from 50 to 65 degrees, my conclusion is that changing the angle is changing when in the day you capture that heat. And my theory is that the higher angle is collecting more in the morning and evening, the lower angle more in the middle of the day. My temptation would be to go with the higher angle for two reasons: firstly, it stretches out how much of the day you can collect for, which arguably will make it more able to capture some heat in short breaks in the cloud; secondly, it means that the unit is less effective in summer. In the middle of summer you’re going to get heat you don’t know what to do with, once you’ve heated your pool you’ll run out of options. You can’t let your system overheat (bad things will happen) so you have to dump the heat somehwere, running a higher angle reduces the overheating problem in summer. My thought is that in summer I’d heat a pool, then dump any residual heat through a coil into the ground under the house. Can’t hurt to have warm ground, there is some information on the web about intraseasonal heat stores (i.e. heating up the ground, and getting it back 6 months later), I reckon it could have benefit.