DIY solar air heating collectors are one of the better solar projects. They are easy to build, cheap to build, and offer a very quick payback on the cost of the materials to build them. They also offer a huge saving over equivalent commercially made collectors.
Two of the more popular designs are the pop can collector and screen absorber collector. The pop can collector uses columns of ordinary aluminum soda pop cans with the ends cut out. The sun shines on the black painted pop cans heating them, and air flowing through the inside of the can columns picks up the heat and delivers it to the room. The screen collector uses 2 or 3 layers of ordinary black window insect screen as the absorber. The sun shines on the screen and heats it, and the air flowing through the screen picks up the heat and delivers it to the room.
This page gives a rundown on building each of the two collectors, and also compares the heat output of the two collectors in a side by side test of the two. The two collectors were built specifically for this side by side test and are identical in size, box construction and glazing -- only the absorbers are different.
Note that for ease of testing these collectors were built small (2 ft by 3 ft), but for real heating you want to go much larger. A 4ft by 8 ft collector is about the minimum to contribute some real heat, and larger is better -- if you can integrate your collector with an entire south wall like this one, that's great.
The test collectors: Pop can collector on left and screen collector on right
There is a lot of not so good information out there on what makes a good solar air heating collector design, so I thought I would include a little info on solar air collector physics, what makes for a good design, and how one can measure and compare collectors accurately. If you are an old hand at this stuff, just skip this section.
On just about all solar thermal collectors, the sun shines through the glazing, and hits the collector absorber heating it. The air flows through the inlet and over or inside or through the absorber picking up heat as it goes. This heated air then flows out the collector outlet and into the room being heated. The main differences between air heating collector designs have to do with how the air flows over the absorber.
In full sun, the incoming solar energy is about 1000 watts per square meter of collector area. Of this 1000 watts/sm, about 10% is absorbed or reflected by the glazing and never gets to the absorber. Of the remaining solar energy, about 95% is absorbed by the absorber. So, for the 1000 watts/sm that arrive at the collector face, about 850 watts/sm end up actually heating up the absorber.
Most of this 850 w/sm that made it into the absorber end up going down one of two paths:: one part is picked up by the air flowing through the collector and ends up heating the room, and the other part ends up being lost out the glazing. The job of the collector designer is to maximize the first part and minimize the 2nd part.
The heat output the collector can be calculated as:
Heat Output = (Temperature Rise)(Airflow)(air density)(specific heat of air)
Temperature rise is the increase in air temperature from the inlet to the outlet of the collector -- often around 50 to 60F for well designed collectors. For example, air might enter at 65F and exit at 120F.
Airflow is the volume of air flowing through the collector expressed in cubic ft per minute (cfm) -- often around 3 cfm/sqft of collector area for well designed collectors.
Air density and Specific Heat are physical properties of air that you don't really have any control over -- air density is 0.075 lbs per cubic foot under standard conditions, and the Specific Heat of air is about 0.24 BTU per lb per degree F.
Its very important to note that the heat output depends on BOTH the Temperature Rise and the Airflow. Many of the videos out there talk only about temperature rise as though that is all that mattered, when it fact its only half the story. It is quite common for a collector to have a very high temperature rise and have a low heat output because the airflow is much to low.
There is a tendency to think that things that increase the collector temperature rise will improve the efficiency of the collector, but, in general, the most efficient collectors will have a temperature rise that is just enough to be used for space heating and an airflow that is relatively large. The reason for this goes back to that portion of the heat that the absorber takes in that ends of being lost out the collector glazing. You want to minimize those glazing losses, and an important way to do that is to keep the absorber temperature as low as possible -- the cooler the absorber runs, the less heat will be lost out the glazing. A way to keep the absorber cooler while extracting the same amount of energy from it is more airflow.
On solar air heating collectors, it is relatively easy to get most of the suns energy into the collector absorber. The difficult part of air collector design is getting the heat transferred from the absorber into the air. Air is a low density material with a low specific heat, and that makes the heat transfer from absorber to air difficult. The things that tend to help in the transfer of heat from the absorber to the air stream are a high volume of airflow, a lot of absorber area, and good and even airflow of high velocity air over the full surface of the absorber. All of these things help to efficiently pick up heat from the absorber, and to keep the absorber at a cooler temperature so that losses out the glazing are minimized.
The good characteristics of the pop can collector from an efficiency point of view are that it has a lot of absorber area (about Pi times what a flat plate would have), and it has a mixed flow of relatively high velocity air through he can columns. The good characteristics of the screen collector are that the thousands of strands of screen wire provide a lot of screen to air heat transfer area, and that the inlet and exit vents are arranged such that the airflow is required to pass through the screen to get from the inlet to the outlet.
While there are no hard and fast rules, a temperature rise through the collector of about 50 to 60F works well in that is is warm enough to feel warm coming out of a heater vent. If the room temperature is 65F, than the collector outlet temperature will be about 120F. Moving air that is much cooler than this will not feel warm. Going for a temperature rise greater than 60F usually means a hotter collector absorber and increased heat loss out the glazing.
Airflow through the collector of around 3 cfm per sqft of collector area for a collector with a well designed absorber is about right. More airflow would make the collector more efficient, but it also increases noise and fan power, and may lower the temperature rise to the point where the air does not feel warm to people for space heating. The about 3 cfm per sqft of absorbers seems to be a good compromise between efficiency and the other factors.
Even though solar air heating collectors have been around a long time, it seems there are still significant design improvements that can be made to both performance and cost/labor of construction. This seems like a very interesting and worthwhile area to work on.
If you do want to work on an improved air collector design you must have a way to measure performance so you know if the changes you are making actually improve efficiency or not. Its fine to speculate on what might help, but if you don't measure the actual performance carefully, you really don't know if a change helps or hurts performance. Right now, when you look across Youtube etc., it seems like we have a lot of speculation and not a lot of careful measurement going on. Its not that difficult or expensive to do side by side tests of collector designs and to measure the performance.
Measuring the absolute performance of a collector is difficult. A collectors performance depends on its design, but is also influenced by solar intensity, ambient temperature, wind and collector orientation -- all things that vary quite a bit from day to day and even minute to minute. One way to get around most of the variations is to test a baseline or reference collector side by side with the collector you are making changes to. If the two collectors are side by side, then they see the same ambient temperature, the same solar intensity and the same wind. If you make a change to your test collector and it performs better relative to the reference collector, than you can be sure the change you made was a good one.
With side by side collectors, you need only measure the inlet and outlet temperatures of each collector and the airflow through each collector. Changes that increase the product of the airflow times the temperature rise improve the heat output of the collector -- its as simple as that. The test setup section below shows a more detailed example of a side by side test. If you set up the two collectors so that you can adjust the airflow such that both collectors receive the same airflow, then the collector with the greatest temperature rise is the winner.
The details on testing below show one set of insturments that allow accurate testing on a fairly small budget. For other low cost instrumentation ideas... But, just buying the cheapest instruments on eaby or at the local store and not doing anything to check their calibration is unlikely to give accurate results.
I'm just going to cover building the pop can and screen absorbers and not say much about the collector box and glazing, as these are the same for both, and are well covered elsewhere. There are some links at the end that also cover building both types of collectors.
The steps in building the pop can absorber -- I've tried to use techniques that don't require much in the way of special tools:
1. Collect the cans.
If you drink pop or beer, this is not really a problem, but if you don't, you need to find a source of intact cans. Recycling places are a place to start. Some civic organizations collect pop cans for recycling and may be happy to let you sort through for the intact ones and sell them to you. You may have friends who will save cans for you.
2. Clean the cans.
The cans typically have some residue and need to be cleaned. Most people just use soapy water for this. Be sure to rinse the soap off.
3. Cut the ends out of the cans.
This requires more skill and effort than it might seem at first. Various ways have been worked out to do this -- I'll show the way I did it, but the links at the end provide other methods.
Cutting out the bottom of the cans:
I tried several methods, and ended up using a spade style wood drilling bit. The bit is 1 1/2 inch diameter. I mounted an electric drill in a woodworking vice, chucked the spade bit, and clamped the trigger on. Then with gloves and safety glasses on take a can in both hands, center it on the space bit, and slowly advance the can into the turning spade bit. This take a little practice, but is not that difficult. But, it is not something for kids to try. There are sharp edges and if the bit catches the can, it can rip it out of your hands easily.
The spade bit I bought (an Irwin) had little spurs out at the edges of the bit. These tended to catch the can suddenly and make the cutting process hard to control, so I ground down the spurs. This makes the process more controllable. A lot of spur bits do not have these spurs at all.
Another method I tried that worked quite well was to use a belt sander to basically sand off the bottom of the can. The can is not flat on the bottom, but has a ridge around it a little ways in from the outer diameter of the can. The idea is to just sand through this ridge, at which time the bottom of the can falls out. Its very fast and controllable. The down side is that it actually takes off a bit of the surface where you will later be gluing the cans together. I decided to go with the spur bit as its almost as fast and leaves a good gluing surface, and does not require any special tools.
4. Cutting out the top of the cans:
First, break the tabs off.
To open the top up, I just used tin snips to make radial cuts in the top of the pop can. Each cut starts at the drinking hole and extends outward toward the can rim. I then bent the resulting pieces out toward can walls by first pushing the can down on a pipe, and then bending the pieces out with a gloved hand.
Again, there are several ways to do this, and the links at the end show some of them. Greg worked out a way to use 55 mm hole saw to cut the lids out.
There may be some advantage to trying to bend the radial pieces remaining after the tin snip cuts in such a way as to cause more mixing of the air, but this would increase the pressure drop through he collector.
5. Gluing the can columns.
To glue the cans into straight columns you need some kind of form to place the cans in while the glue sets. I used a couple of boards nailed together to form a "V". Greg used a length of PVC pipe that was sawed in half.
Put the first can in the form with the top pointing up, apply a bead of glue around the bottom of the 2nd can and press it down onto the first can. Repeat this process until all the cans for one can column have been glued. Place a weight on the top of the column to maintain a little pressure while the glue sets.
For glue, I used silicone caulk. Its a good high temperature material that remains a bit flexible and does not stink or outgas after its initial cure.
Keep in mind that the cans must not only be glued together but also sealed so that air does not leak out between the can joints and into the collector box.
6. Preparing the cans columns for painting:
The can surfaces should be prepared for painting. I just wiped them down with acetone, but a more careful preparation may be in order to insure the paint stays well attached over the long haul. Some of the links at the end may have better preparation methods.
7. Build the manifolds:
There are manifolds along the top and bottom of the collector that distribute the air evenly to each can column. The manifold wall also holds the can columns securely in position.
I used thin plywood for the manifold wall. A 2 3/8 inch hole saw turned out to be a good fit for the cans to nest into at both the top and bottom. Its important to take some care in laying out the holes that accept the cans to get an even spacing. I taped the two manifold boards together and drilled both at the same time -- this only works because the 2 3/8th hole size fits both the top and the bottom of the cans snugly.
While the painted plywood manifold walls may have an acceptable life, and manifold wall made from sheet metal would be a better way to go.
8. Install the inlet and outlet vents:
Cut holes in the back of the collector for the inlet and outlet vents. I used round duct "starter collars" made for starting a duct run from a larger duct. I used Great Stuff polyurethane foam to seal and bond the starter collars in place.
9. Assemble and paint the absorber:
Use silicone caulk to glue the manifold walls to the can columns.
Paint the absorber assembly with a high temperature black paint. I use Rust Oleum flat BBQ paint in a spray can. All of the inside of the collector box and manifold boxes should also be painted black.
Mount the absorber assembly in the collector box and make sure the manifold areas are sealed up so that airflow does not leak from the manifolds into the collector box. Let the completed collector box and absorber sit out facing the sun without the glazing on a good sunny day to bake out any volatiles.
To cover the front face of the manifold, I used a piece of clear Plexiglas so that the sun shining on the manifold area would be absorbed by the black back of the manifold box and have its heat recovered. Its important to seal the manifold box carefully as you don't want air leaking from the manifold into the rest of the collector box. For long life in this high temperature environment, polycarbonate with a UV resistant coating would be a better choice than the Plexiglas (acrylic).
1. Build the absorber frame:
The absorber frame holds the three layers of screen in place.
Using 1 by 1 wood (actual dimensions 3/4 by 3/4 inches) make a rectangular frame that will just fit inside the collector box.
2. Staple screen to frame:
Start with the middle layer of screen. Cut out a piece of screen the size of the outside of the frame. Staple this piece in the frame so its about half way between the top and bottom face of the frame. No need to be very exact about this. Put a little tension on the screen as you staple so its not to baggy.
For this collector, I used fiberglass screen. Black aluminum screen might be a better choice for long life. I did have one instance in the past when the fiberglass screen (or the coating on it) outgassed and deposited a film on the inside of the glazing. I have not seen that on this fiberglass collector. The thermal performance of the fiberglass and aluminum screen appears to be quite similar.
Do the top and bottom layers of screen as one unit. Cut a single piece of screen that is big enough to wrap around both faces of the frame -- so, if you had a 4 by 8 ft frame, the screen should be 4 by 16 ft. Lay the screen over one side of the frame and staple it using a little tension. Flip the frame over and wrap the rest of the screen over the bottom of the other side of the frame and staple it.
Stapling the middle layer of screen in
Rethinking this, I think that it would likely work fine to just staple the three layers of screen directly to the sides of the collector box. This would save making the frame to hold the screen. I would make a few guide marks at the top and bottom of the collector box, then staple the screen across the top, then pulling the screen tight, staple across the bottom, then putting a bit of tension on the screen staple each side. Remember that the screen needs to be close to the back of the collector on the inlet end and close to the glazing on the outlet end. About a half inch space is normally allowed between each layer of screen, but I don't believe that anyone has done any testing to see what the optimal spacing is.
3. Install the inlet and outlet vents:
Cut holes in the back of the collector for the inlet and outlet vents. I used round duct "starter collars" made for starting a duct run from a larger rectangular duct. I used Great Stuff polyurethane foam to seal and bond the starter collars in place.
4. Mount the absorber frame in collector box:
Place the frame in the collector box. It should be tilted so that at the inlet end, its near the bottom of the collector box, and on the outlet end it should be near the top of the collector box. Secure the frame to the box with a couple screws on each side that extend through the box and into the frame. Seal any gaps between the edges of the absorber frame and the collector box with something like Great Stuff foam or silicone caulk.
Absorber mounted in collector box. Note that the absorber is tilted so that it is close to the bottom of the box on the inlet end, and close to the top
of the box on the outlet end.
Screwing the absorber frame in place through the side of the
side of the collector box.
5. Inlet to absorber connection:
On the inlet end, press the three layers of screen down onto the inlet flange and glue them in place with polyurethane glue. Be sure to work the glue down through the layers of screen so that there is glue between the bottom layer of screen and the starter collar metal tabs. Weight down the screen until the glue sets. Use a piece of waxed paper between the weights and the screen so the weights don't get bonded to the screen. After the glue sets, cut out the screen inside the inlet duct. The idea is to introduce the inlet air to the space between the screen and the glazing so that the air has to flow through the screen to get to the outlet vent.
Finished inlet --with the layers
of screen cut out over the inlet duct.
With this arrangement, the air is
introduced between screen
Rethinking this, it seems like it would make more sense to 1) cut the hole in the back of the collector for the inlet vent, 2) install the screen absorber, 3) cut an opening in the screen matching the inlet vent hole in the back of the collector, 3) apply Great Stuff to the inlet vent (which is a start collar) and push the inlet vent down through the screen and the hole in the back of the collector. This (I think) would be simpler and more secure than the method I used on the prototype.
5. Install the inlet baffle:
A baffle is used across the bottom of the collector to spread the inlet air more evenly across the width of the collector and to keep the inlet air from impinging directly on the glazing (which would increase heat loss). The ideal flow is for the inlet air to spread out evenly over the screen, and then flow through the screens toward the back of the collector picking up heat from the solar heated screen layers. This baffle is made from some aluminum soffit material I had on hand -- it is 0.018 inches thick. Aluminum flashing painted black would likely work as well.
The two finished collectors setup for testing.
I wanted to check and see if each of the columns of cans was getting about the same airflow.
With the glazing off, and the top manifold cover off, I hooked up the lower inlet manifold to the blower and measured the air velocity in each can column. Ideally all of the air velocities would be equal.
The table below shows the measure air velocity in each can column in feet per minute.
There is some variation, but overall, I'd call it a good distribution.
There is no apparent pattern to the variations and I suspect that the variation is due to some can columns having more air resistance than others just due to the variation in the way the top of the cans were cut open?
I did a test some time back on a downspout collector, which is similar to the can collector except that the aluminum downspouts are used instead of columns of cans. The downspout collector showed much more variation in velocity between downspouts. I think that the difference may be that the can columns have more resistance than the downspouts, and this allows the manifold to act more like a constant pressure plenum.
The two collectors (each 2 ft wide by 3 ft high) were mounted on a stand that holds them at the same orientation to the sun.
Each inlet duct was connected to its own small blower. The blowers are duct booster fans sold at Home Depot. They are not very good fans when there is any significant pressure drop, but the fact that both the pop can and screen collectors both have low air resistance and only 6 sqft of area made them just about right for this job.
The two identical inlet blowers.
The shutter that is used to adjust both collectors to the same
The back of the collector stand showing the inlets and blowers for each collector. The inlet to each blower has a shutter that allows airflow to be adjusted to provide the same airflow on each collector.
The collector outlets each have about 2 ft of 4 inch duct attached to them to allow the airflow to settle down and become more uniform. Airflow is measured at the end of the duct with a calibrated Kestrel turbine style anemometer.
The two 4 inch outlet ducts at the top, and the
Kestrel anemometer measuring outlet flow.
For this test, the outlet temperatures were measure with Hanna Checktemp thermometers. These thermometers come with a calibration certificate, have an accuracy of 0.3C, and have built in calibration check. Inlet temperature was the same for both collectors and was measured by a Taylor alcohol thermometer placed between the two collectors at inlet level.
The Hanna thermometers are next to each other at the top, and their stainless steel probes are the
The two Hanna thermometers.
Cardboard shading disks were put on top of the glazing to shade the outlet ducts
from direct sun and avoid direct sun on her temperature probes.
I did a tests on a couple days with very similar results -- the results below are for the 2nd day (7/21/13).
Both were very sunny days with no visible clouds or haze, and light to no wind.
The ambient temperature was mid 80'sF at the time of the test. Not ideal for testing a space heating collector, and I will repeat the test when we get some cold weather.
For the test, the outlet duct flow velocity for both collectors was adjusted to 235 ft per min. This is the velocity of the pop can collector with the flow adjustment shutter all the way open. The screen collector has a bit less flow resistance and the shutter was closed part way (as shown in the picture above) to cut its velocity down to 235 fpm. The area of the 4 inch duct is 0.0872 sqft, and the resulting flow rate is (235ft/min)(0.0872 ft^2) = 20.5 cfm, or 3.4 cfm per sqft.
For the 7/21/13 test:
|Time||Pop Can Collector||Screen Collector|
The Heat Out column is computed as:
Heat Out = (Trise)(Flow)(Density)(Specific Heat)
For example: At 1:40 AM Heat Out = (61.3F)(20.5 cf/min)(60min/hr)(0.061 lb/cf)(0.24 BTU/lb-F) = 1104 BTU/hr (or 184 BTU/hr per sqft of collector).
The air density is low due to our 5000 ft elevation (sea level standard air density is 0.075 lb/cf).
So, for this test the heat output from the screen collector was about 6% higher than for the pop can collector. This may change for more typical winter heating conditions.
Update: I did repeat some of the performance testing in winter conditions. The performance of the two collectors was closer and nearly equal.
At one point I noticed a slight filmey coating had appeared on the inside of the pop can collector glazing. I took the glazing off and cleaned the film off, reinstalled. After a few minutes, the temperatures settled back in to where they had been, so the film was not opaque enough to make any difference. Not sure what caused the film -- perhaps outgassing from the paint.
At one point I reversed the two thermometers just as a check on their calibration. After a couple minutes, they settled down to very close to the old readings.
The IR (thermal) image was taken at 11:00 am. It shows the temperature of the glazing on each collector. This is important because most of the collector heat loss is through the glazing, and cooler glazing is an indication of more efficient operation.
Thermal image of the two collectors: pop can on left and screen on right.
The screen collector shows a lower average glazing temperature, which does tend to agree with it higher heat output. The screen collector glazing temperature is fairly even, which might indicate that the picture of the inlet air spreading over the full surface of the screen and then flowing through it might be true. The pop can collector glazing shows a smooth increase in temperature from top to bottom with no apparent hot spots that might indicate uneven flow through the can columns.
This IR picture shows the pop can collector with the glazing removed:
This is showing the actual temperature of the surface of the pop can absorber. It is nice and even with no apparent hot spots that would indicate uneven airflow.
The absorber temperatures seem (to me) to be on the high side given that the inlet air temperature was about 85F. For example, on a good water heating collector, the absorber temperature near would only be about 10F over the water temperature. This might indicate that steps to improve the heat transfer from the cans to the air would payoff?
Anyone glean any other conclusions out of these pictures?
One of the advantages of the pop can collector is that solar radiation striking the front of the can is transferred through the can walls around to the back of the can, and this provides more heat transfer area for the can to heat the air flowing through it. But, this is only true if the the heat is actually transferred around to the back of the can. On this test, I put thermocouples on the front and back of one of the cans to get an idea how efficiently the heat gets around to the back of the can. If the can walls were a perfect heat conductor, and front and back of the can would be at the same temperature and both front and back would be equally effective in transferring heat to the airstream.
<pics of thermocouples>
Here is measured data
|Can Temperature||Air Temperature||Difference|
|Front||162 F||84 F||78 F|
|Back||122 F||84F||38 F|
The heat transfer between the can wall and the airstream is proportional to the temperature difference betwen the wall and the air. This difference is 78F on the front wall of the can and 38F on the back wall of the can. So, it looks to me like the back of the can is only about half as effective as the front of the can in transferring heat to the airstream?
The table below gives a cost estimate for the two collectors in a 4 by 8 ft size and assuming a simple box construction and using SunTuf corrugated polycarbonate glazing. The box costs are the same for each.
|Item||Pop Can Collector||Screen Collector|
|Glazing -- Suntuf corrugated polycarbonate||$44||$44|
|Glazing end fillers||$7||$7|
|Back - 1 inch polyiso rigid foam board 4 by 8||$21||$21|
|Starter collar ducts||$4||$4|
|pop can absorber|
|Manifold top and can wall||$10|
|Screen 4 by 24 ft||$14|
|absorber frame (if used)||$3|
|Miscellaneous paint, fasteners, Caulk, GreatStuff,...||$25||$25|
These prices are from my local Home Depot and will vary from place to place, but it gives an idea.
This does not include any ducting, blowers, or controllers which might well double the cost.
The pop can collector saves a little money because of the free cans used for the absorber.
Both collectors probably have room for improvement in the designs.
The pop can collector:
Comparing the two collector designs in several categories:
A clear advantage to the screen collector.
Building the boxes takes the same amount of time, but the time to construct the pop can absorber and manifolds is much greater than the time to construct the screen collector absorber and inlet baffle.
About an 5% advantage to the pop can collector.
Under winter conditions, the two collectors perform nearly identically.
I would expect both collectors to have good lives if built carefully and with good materials.
Some links on building solar air heating collectors:
Questions? Comments? Suggestions? -- this is the place.
blog comments powered by Disqus