Solar thermal collector autoconstruit: what solution?

[cortejuan]

Hi,

before posting, I read the various threads on the subject. Those are my questions.

- I recovered new double glazing and low emissivity, I have others that are about 20 years old. Which to choose as the front face of my future sensor, isn't low emissivity a handicap?

- I have three solutions to make the sensor:

- I use a radiator in aluminum profile (which I own), I paint it black and I integrate it in an insulated box with glass wool.
- I use one or two car radiators (which I own) and which I also smear with black.
- I make a sensor out of copper pipe screwed on a steel or aluminum plate.

All this to heat a water reserve of one cubic meter to heat my old greenhouse under renovation. The greenhouse is 3,5x2m and is made of glass.

Same principle as the installation of a new greenhouse which I mentioned on another thread "heating a greenhouse by thermal buffer".

I look forward to your advice.

Ah yes, telling me a host of sites to consult does not interest me it's already been done ...

cordially

[BobFuck]

As the T ° of the water will be low it is not necessary to insulate very strongly or to seek the sensor with high output, but rather the least expensive.

For example, a solar pool heater (finds it the cheapest possible) behind a glass with a PIR foam panel behind.

http://www.maison-facile.com/boutique/b ... 4686&b=204

> isn't low emissivity a handicap?

no, but you better put them on the greenhouse rather than on the sensors!

[chatelot16]

to make domestic hot water, double glazing is useless: 2 layers of glass produces 2 times more loss by reflection ... double glazing could reduce heat loss by convection, but a single glass is enough

double glazing therefore makes more loss by reflection than gain by avoiding heat loss

for pool heating it is worse: we only need it when it is not very cold and we want to heat at very low temperature: we recover more power with a sensor without glass at all

to heat a greenhouse in the middle of winter, we also want a low water temperature but it has to work when it is cold: I see simple glazing, not even necessarily glass: it could be in film for greenhouse, precisely because that it will never heat up like a sensor for domestic hot water

why not use the surface of the greenhouse? because it already naturally serves as a solar collector: the advantage of adding solar collectors is to make the surface extra, which brings in heat because the pump is turned on when necessary. and which never loses heat because the water circuit is closed when the sensor is cold

close when the sensor is cold, it may freeze, even better plan the sensor so that it empties completely when it is not in use

[cortejuan]

Hi,

thank you for this information even if it does not go in the direction that I hoped. Regarding the use of the greenhouse as a sensor, it is done and it works rather well with my second greenhouse. The only problem is that I would like to increase the efficiency in heat capture with a more efficient system. I thought that by using a specific system I would bring in more calories when the sun was beating down. But maybe I'm wrong.

For example, right now, I am gaining two degrees per day on 2800 liters of water between morning and evening in slightly sunny weather. Which makes me roughly 6,5kWh available for the night. I would like to increase the amount of energy stored for very cold days because currently, I am not autonomous with my only storage, I have to bring some energy via a propane gas system when the night temperature is around 0 degree.

cordially

[chatelot16]

to make an effective solar collector in winters it must also be effective fairly quickly: it should not lose 2 hours before starting to heat: therefore sheet metal plate with a 6mm copper pipe welded in zig zag above ... especially no too massive recovery radiator too long to heat

[BobFuck]

By the way, how many m² do you have for the sensors?

[dedeleco]

to make an effective solar collector in winters it must also be effective fairly quickly: it should not lose 2 hours before starting to heat: therefore sheet metal plate with a 6mm copper pipe welded in zig zag above ... especially no too massive recovery radiator too long to heat

Yes, very exact, and read on wikipedia diffusivity and thermal conduction to understand thermal inertia.
In winter it is necessary to insulate well, hence under vacuum a little better.

See the APPER site full of information, advice and achievements.

In a simple sensor, very expensive copper pipes are useless, a simple plastic hose of 16mm watering wound at 20 € per 100m is enough, and works as well, because the enormous thermal conductivity of copper is useless in this case, if we think after reading the basics of thermal conductivity.

[chatelot16]

copper conductivity is useless: plastic can be enough ... but the platic does not weld on the sheet: you have to put a surface of pipes equal to the sensor surface ... it takes a long time to heat up

with galvanized sheet of 0,5mm and a welded pipe every 5cm is not bad

I also thought of a sheet of aluminum folded every 10cm with a shape suitable for wedging a 16mm polyethylene pipe ... but it requires an industrial folding means ... the good old copper tube welded to the tin on the sheet remains the easiest way

[dedeleco]

In my opinion, the thermal conductivity of the air inside is sufficient, seamless, with black plastic, which you can calculate with thermal convection (ie 5 mm of still air), with 16mm pipes close enough (see unpretentious sensors and APPER !!).
All this can be measured in simple tests to avoid making it complicated and almost useless, more expensive !!!

[BobFuck]

I asked you the surface you have, because you have several options. There is a compromise between the cost of the sensor (say m²) and its performance.

At constant budget, if you have a lot of space, a really cheap sensor per m², with a terrible yield, can produce more than a sensor more expensive per m² and more effective, but smaller.

On the other hand, if you have little space, you have to work more on the yield ...