This is a small test of two types of ordinary insect window screen used as the absorber in a solar collector. Motivation for the test is just to understand better how much the screen absorbs and let through, and to think about ways to improve the performance.
Two layers of window screen appear to do well as the absorber for solar air heating collector. Here is one example... In the testing that Scott and I have done to date, the two layers of screen have done about as well or better than other more complex absorber configurations tried so far.
Typically in these collectors that use screen for the absorber, the intake air is introduced across the bottom of the collector on the glazing (sun) side of the screen absorber. The outlet vent is at the top of the collector at the back of the collector box such that the air has to flow through the absorber from the glazing side to the back side to get out. The thinking is that the sun heats the black screen, and the air flowing through the screen picks up the heat. The screen is thought to work well because it provides a lot of surface area for heat absorption, and because the screen has a small amount of air resistance that helps to spread the air evenly over the full absorber.
A somewhat different alternative on how the screen works was proposed by Laren on the SimplySolar Yahoo discussion group last week. This is that the screen has so much open area that 2 layers do not absorb a major part of the solar, and that a lot of the light gets through the screen and is absorbed instead by the back wall of the collector. The screen absorbs some light, but also acts as a barrier to keep the heated air behind the screen layers from moving forward and contacting the glazing -- which would result in high heat loss. It seems plausible to me that this is at least part of what is going on, and one clue to finding out which model is right is to determine how much sun the screen actually absorbs.
In order to try to get a better idea how much of the solar is absorbed coming through the screen, vs at the back of the collector box, I did the small experiment described below. Basically, I just measure sun intensity with a pyranometer first without any screen, and then with 1, 2, and 3 layers of screen in front of the pyranometer. The pyranometer reading should drop by an amount roughly equal to the amount of solar absorbed by the screen(s).
I tried fiberglass and aluminum screen -- both black.
Looking through 2 layers of each kind of screen.
Looking at the sun through 2 layers of alum screen
Looking at the sun through 2 layers of FG screen.
Apogee pyranometer without the screen in
front of it.
2 layers of FG screen in front of pyranometer.
The wire diameter on the FG screen was 0.008 inch, and the thickness of the screen is 0.0125 inch.
The wire diameter on the alum screen is 0.009 inch, and the thickness of the screen is 0.021 inch.
The table below shows the sun intensity readings through 1, 2, and 3 layers of fiberglass screen, and through 2 layers of black aluminum screen.
The percent of light absorbed is also shown.
The logger plot that the numbers numbers in the table were derived from is shown at the end of the page. Three runs were done with 2 layers of FG screen and 2 layers of AL screen, and 2 runs with 1 layer of FG and with 3 layers of FG. The fraction of sun light getting through is calculated separately for each run using the average of the full sun values just before and after the screen is placed over the pyranometer. The full sun value is divided into the reading with the screen in place to estimate the fraction of light that the is getting through the screen. Then the absorption readings for the runs are averaged to get the number in the table. As the plot show, the variation was pretty small. It was a good sun day -- some clouds away from the sun, but none in the vicinity of the sun. It was windy and about 46F.
I held the screen about 3 inches in front of the pyranometer head by hand. The idea of holding the screen by hand is that it makes the screen shadow pattern on the pyranometer head move around --so, if the reading changes as the shadow pattern moves, holding the screen by hand should help to average any variation out -- doing 3 runs also helps average things. The readings from the different runs agreed well, so I don't think that the shadow pattern is an issue.
|Fraction of sun getting through to pyranometer||Fraction of sun absorbed by the screen(s)|
|1 layer FG screen||0.56||0.44|
|2 layers FG screen||0.34||0.66|
|3 layers FG screen||0.21||0.79|
|2 layers AL screen||0.48||0.52|
So, it looks like 2 layers of the FG screen absorbs about 66% and 2 layers of the Al screen absorb about 52%.
- A fair bit of sun does get through the 2 layers of screen, and the back wall of the collector does apparently play a significant role in absorbing the sun.
- At least for these two samples of screen, the fiberglass screen absorbs somewhat more sun than the Aluminum screen.
- Each layer of screen added absorbs less light - basically, if you start with 100 units of light, the first screen absorbs 44 units, adding the 2nd screen absorbs 22 more units, and adding the 3rd screen absorbs 13 more units. This is what you would expect as the front screens partially shade the back screens. So, each layers is less effective at absorbing but does still absorb some more light, and also provides more heat transfer area.
- You have to wonder if adding the 3rd layer of screen to move more of the absorption up to the screen plane would not be better? (see note 1 below) .
- If the back wall of the collector is doing a significant fraction of the absorbing, maybe some changes could be made to improve its absorbing characteristics -- like more surface area?
- The back wall of the collector is going to be running hot. It seems like it wants to be well insulated to reduce heat loss, and should be isolated from high thermal mass items (like the back of the collector box).
- If the back wall of the collector were 100% reflective, and the screen had the same absorption characteristics on the way out, then 2 layers of FG screen would let out (0.34)(0.34) = 12% of the incident sun light, and 3 layers with a reflective back would let out 4% of the sun light.
With a 90% absorption collector back, 2 layers would let out (0.34)(0.1)(0.34) = 1.1% of the incident sunlight, and 3 layers with a black back would let out 0.4%.
Anyone see anything wrong with the way the way all this was done? Or, with the conclusions?
Anyone have any absorber improvement ideas?
The pictures below show the shadow pattern on a white board for various screen configurations.
Three layers of fiberglass screen:
Shadows from 3 layers of FG screen.
Two Layers of FG screen:
Shadow from 2 layers of FG screen.
One and three layers of FG screen:
This is the frame that holds the 2 layers of FG
screen with one layer folded back so the left half is
3 layers of screen and the right half is one layer.
One layer shadows on right and 3 layers shadows on left.
This is the shadow pattern on the pyranometer head for 3 layers of FG screen.
Fiberglass screen close-up:
Lots of open area.
Based on a blow up of the picture above, I'd estimate the open area at roughly 62% .
The pyranometer says that 56% of the light gets through one layer of screen, so the percentage of absorbed light appears to be a bit higher than the percentage of screen open area.
The logger record of the measurements is shown below.
Note (1) -- I did try 1 layer vs 2 layers vs 3 layers on my thermosyphon collector, and in this case 2 and 3 layers about tied and both better than 1 layer. But, on a fan forced collector 3 may be better than 2?
Gary Feb 15, 2011