Introduction: Screened Solar Air Heater


I've been interested in Solar projects for a while and several years ago I ran across an "Aluminum Can Solar Air Heater". I thought it was a great build with some scrap materials and vowed that I would one day build my own. After all, I have a large, unheated space that would greatly benefit from a few free BTUs this Winter.

I researched several builds, explored what materials I could grab locally and figured it was time I built my one myself. And then I had time to think about it a bit....

Step 1: The Concept

The idea was pretty straight forward.

"Build a wooden frame, insulate with foam-board, paint the inside black, use screening to capture the heat and then cover everything with an Acrylic sheet. The idea is simple. An enclosed box, insulated against the cold, takes air from indoors and heats it up by absorbing energy from the sun. Once heated it passes the warm air back indoors.".

What could be easier? What we ended up with was indeed a wooden frame but we also incorporated (2) types of foam board, a concrete backer board as a thermal sink and an overly complicated way of mounting the screening.

Original designs I've found on the Net incorporated sleeves of aluminum cans and some sort of baffle. The cans are painted black and as air passes up one sleeve and down the other it is warmed before being returned. This design worked well but there was a question about surface area and heat transfer. One of the issues that I read about this design is that there was a large air-space between the glass and the tops of the cans that didn't mix with the airs going through the cans which wasted a lot of thermal potential.

I then came across a few builds that suggested a simple charcoal "screen", angled across the collector, would work as well or better because it had a larger surface area to interact with the moving air.

I of course believed this to mean that more is better so I came up with a wave patter with the screen. The screen would wave up and down, top to bottom, in order to increase the solar absorption area as well as the surface area interacting with the air.

I also wanted to include some sort of heat sink. The thought process went along the lines of being able to keep radiating heat to the air in case there was a momentary drop in sunlight. I first thought of bricks, rocks and other such materials but all were too heavy on a gut-guess on how well it will benefit the overall output.. I opted for a thin Concrete Backer-board. If it worked, great. If not, not a lot wasted.

Lastly, and for no good reason, I decided I wanted air channels behind the collector. The reason was to "pre-warm" the incoming air against the back of the thermal sync before being introduced into the main heating area. I figured the back of the board would be pretty hot and why waste the heat, why not run some air up against it and pre-charge it a bit. Again - NO IDEA AT ALL if that's actually happening and even if it is, if it's worth the design hassle. My guess, once everything is finished, it may behoove me to build the next one without this feature. Simpler, easier, and a deeper area for additional screening material.

Step 2: Overthink, Overdesign, Overbuild and Learn From Your Mistakes

Besides the part where I'm just staring off into the distance thinking about a design, this is where the project actually starts. I'll sit down with some paper and start sketching out my thoughts. I find this a good way to work out a design, assembly steps and materials list.

While drawing up designs I'll go through many, many iterations of the design which helps make the final build go quicker and smoother. However, there's a dark side to this process. The constant redesign. Noooooo!

Give me enough time and I'll always find ways to make things "better". Yeah, better. This usually takes the form of some kind of over-complication, special materials needs or a build process requiring one-off specialized tools. It's a weakness but it's how the gray matter works. I'll finish a design, all proud and happy, and then I'll hear a whisper of my Dad's voice, "simple is elegant, simple is clean, why so many complications?".

Soooo, this may be one of those not so simple designs. It took WAY too much time between the design, and re-design and several other re-designs until I got the materials and started banging things together. That being said, I enjoyed every step of it and when I make the next one I'll incorporate what I've learned from this one.

My suggestions, for what they're worth:

  • Do work on the design until you're happy.
  • Do not loose your mind in the process, this isn't life and death.
  • Simple is many times better than complicated although sometimes it can't be helped.
    • (However, keep thinking about a more simple design, inspiration strikes out of the blue if kept on the back burner)
  • Everything is a learning experience, good, bad or "Quick, call the Fire Department !"
  • Take what you've learned and apply it to your next build.
  • Always have a next build waiting even if it's just a thought that distracts your from your day.

Step 3: Build 1 - the Wooden Frame


  • 1x6" Pine
  • Kreg Pocket-Hole Jig
  • 1/4" Plywood
  • Acrylic Sheet

The first piece I bought for this project, before I even really put any thought into anything was the Acrylic sheet. It was 36"x72"x1/8". Naturally, this is what determined the size of the frame.

I made the frame itself about 1/4" larger than the acrylic sheet so I could router the edge. I wanted the sheet to drop in flush with the edge of the wood.

  • The top of the frame was routed down about 1/8" for the sheet and the bottom of the frame was routed down 1/4" for the plywood bottom.

I used the Kreg jig to drill screw holds to hold the frame together. I also used some construction adhesive for good measure.

NOTE: Additional builds will have the boards fit together using a dove tail jig. It's a tighter and more structurally sound joint than screws and glue.

Once squared, I dropped in the plywood bottom and pipe clamped the unit together. The plywood bottom was secured with the same adhesive and some small brads. It fit flush to the bottom of the frame due to the routed edge.

I also took the opportunity to cut out two 4" holes in the plywood, towards what would be the bottom of the frame. These were for the in and out airflow. At this point it was an estimated guess, based on my drawings, of where they'd go and how far apart they should be.

Step 4: Build 2 - the Foam


  • 1/4" Foam Board
  • 1-1/4" Foam Board
  • Silver HVAC Tape

"Why two types of foam?"

Excellent question. Strap in for a mediocre answer. I used 1/4" foam board to insulate the bottom and sides of the box. The 1-1/4" foam would also be used to insulate the bottom and sides but would also be used as a structural support for the backer-board and to create the waves in the screening material.

NOTE: I used a utility blade to cut the thin foam but had better luck using a small toothed wood saw to cut the larger foam. I also kept a vacuum (must have) on hand with a soft bristle brush tip to suck up the millions of small pieces it kicks up. I did use a long blade on a few pieces for the screen angles but they ended up not being as square as when using the saw.

I used construction adhesive to secure the foam to the wood and the foam to the foam.

Two other reasons that I used two layers of foam were because of the backer-board and the air vents. The backer-board rested on top and below the thick foam and butted up against the thin foam. I didn't want it to but up directly against the wood as I wanted it to hold onto as much heat as it could. The other reason is the air vent. It had a lip that was sandwiched between the two sheets to help it say put and airtight.

The thin foam didn't lay flat so I ended up using some weights I had laying around to keep it flat until the adhesive dried. I also used a bead on all the edges to help keep things sealed up.

Oh yeah, and don't forget, DO NOT jab your finger with the knife and bleed all over everything.

Step 5: Build 3 - Vents and Air Channel


  • 4" Aluminum flange tubes
  • Silver HVAC tape
  • Aluminum Foil
  • Construction Adhesive

This step was very labor intensive and a bit of a pain. I had to cut several 1" strips from the large foam sheet which required a substantial amount of vacuuming up all the detritus. If that wasn't bad enough I had the hardest time getting the backing off of the silver tape. Lots of pieces of tape, low light and bad eyes - I have to keep reminding myself that this is enjoyable!

NOTE: I'm not completely sure that the air channel idea is completely worth the time it took. It sounds good in my head but perhaps someone with better engineering knowledge could help me out with that decision.

I started by taping up the two 4" vents into the foam. This created an airtight seal and covered the exposed edges of the foam.

NOTE: I wanted all exposed edges of the foam covered. I believe they'd deteriorate over time if left exposed.

I then cut a bunch of 1" strips from the large foam board and constructed the air channel as well as lining the edges. I used the adhesive to glue down the strips and silver tape to cover the exposed edges of the strips. I also used the tape to seal the gaps between the strips and the foam panel.

The backer board will have vent holes to allow the air into and out of the channel so I widened the top and bottom channels to allow for those vents.

Lastly, I created an aluminum sheet (out of several pieces) and adhered it to the top of the channel. I was concerned about the untreated side of the backer-board. I didn't want it to shed material into the air stream.

NOTE: Yeah, so perhaps by restricting the airflow to a narrow channel I'm reducing the intended transfer ability. If I do keep this feature in the next build I'll allow for a full sheet transfer of air with a few support strips to not only help support the backer-board but also help agitate the air and prevent an smooth flow.

Step 6: Build 4 - the Heat Sync


  • (2) Concrete Backer Board 36" x 60" x 1/4"
  • Construction Adhesive
  • Silver Tape

The purpose of using the concrete backer board was to act as a heat sink and help even out the flow of heat when a cloud floats by. I was also hoping that it might extend the wind down of heat when the sun goes down.

NOTE: Perhaps, if this is indeed a good idea, a thicker/denser material might be needed. My fear is that the 1/4" board is too thin to have much of an impact. Or perhaps not.

After measuring for the vents I used a drill for some starter holes and then used a jig-saw to cut out the vents. I used silver tape to cover the edges and adhesive to seal the seams.

NOTE: I read somewhere that some backer boards have fiber issues and may cause health issues so I made sure I coated all exposed surfaces with either tape or paint.

NOTE: Again, if I do an second build I'll expand the vent opening the full width of the board.

I used adhesive to secure the board to the underneath foam strips

Step 7: Build 5 - Paint It Up!


  • Black Oil Based Exterior Paint
  • Matte Black Hi-Temp Spray Paint

I wanted the exterior of the box to stand up to the elements so I opted for an Exterior oil based paint. I put on a couple of coats to make sure that the rain and snow would have (fingers crossed) minimal impact.

I also sprayed the backer-board with hi-temp spray paint to not only absorb the sunlight but to also seal the fibers.

NOTE: Next build I might opt for a brush application for the matte black. I had a few instances where the spray paint

I also painted all other exposed surfaces with the matte black hi-temp paint as their pieces were sized and cut.

Step 8: Build 6 - the Screen, Do the Wave...


  • Black Fiber Screen - 36"
  • Thin Copper Wire
  • Screws / Staples
  • Adhesive
  • Silver Tape
  • Matte Black Hi-Temp Paint

NOTE: I used charcoal fiberglass screening for this project because I had a roll left over. However, I was thinking that it may be a good idea to use the charcoal aluminum screening. Less forgiving if bent but better heat transfer and possible longevity.

The first step was cut the the thick foam board to form inside borders of the form so they rest on top the backer-board but are EVEN with the bottom routed edge at the top of the boards. Make sure to mark which is which, trust me on this one.

I then cut the side strips into triangles. The "bottom" triangles are the ones with their flat side attached to the backer-board. These I just painted black. The ones that have their flat side facing the plexi, I covered all exposed sides with silver tape and also painted black.

NOTE: I don't want to harp on this but DO NUMBER THE PIECES. Yes, you will mix up the order, yes it does matter, yes it will cause you a certain amount of grief and frustration.

The original thought was that the foam triangles would be enough substance to force the screening into waves. This was quickly proven to be an incorrect assumption. A quick scavenge of available items and I had a spool of copper wire and some small screws. I used a balsa saw to cut a small slots where each screen peak would be. On one side I screwed down the wired and then pulled it tight-ish. I wrapped the wire around another screw and tightened. As the screw turned it tightened the wire.

NOTE: Do not tighten the wire too much. As the screw itself turns it continues to wrap the wire. Too much pre-tension and the wire will snap immediately. Too much and it will snap soon after. It needs to be just enough to hold up the screen, no more. Once I snapped a few it wasn't too hard to judge.

I laid out the screen on the floor and cut it to fit inside the form. What I would do next time is make sure there was NO foam pieces on the floor first. Wow does this thing attract foam like a magnet. Static charge my arse.

After the wires were installed the screen actually installed without too much fuss. I stapled the top part of the screen and then used mild tension as I created the first swell. I used a decent amount of adhesive on the first triangle and pressed into place. The adhesive bleeds through the screen and locks it into place.

NOTE: I used long push pins to hold everything in place as it dries. The pins not only keep the foam in place but also hold tension on the screen.

Left, right, top to bottom, each of the triangles drop into place and if you're lucky everything curves up and down without problems. The wires did an excellent job of holding the top curve form. I stapled the screen to the bottom just like the top.

Measure, test fit and then measure again. I thought I had measured and tested well but the one side of triangles came up just a skosh too high. You'll see what I had to in the next step to compensate.

Step 9: Build 7 - Glass Up the Place


  • Acrylic Sheet
  • Construction Adhesive
  • Screws/Washers
  • Silver Tape

I had originally test fit the acrylic sheet when I framed out the box so I was fairly confident it would still fit like a glove. Wow was I wrong and the more I thought of it the more I should have anticipated this problem. The issue is that as everything dried, the tops of the boards bowed in slightly.

NOTE: The top of the box could have been kept in proper alignment and if I had installed a couple of pieces of angled steel. They would have worked well with the curved swells of the screening and kept the sides from distorting.

Well, lessons learned. I used my Dremel to cut down the sides of the sheet so it would once again fit into the routed edge of the frame. I used a few of those cutting wheels but next time I'll opt for the wheel with the little teeth. The high RPMs caused the cutting wheel (toothless) to just melt the sheet which is not an optimal solution.

Yes, the sheet fits gain! NO, THE SHEET DOESN'T FIT AGAIN! Yeah, this is where the problem of misalignment comes back into play. The one side of the triangles was a bit higher than it should have been and the sheet popped up from the frame. I tried to crush the foam a bit but alas it was not to be.

NOTE: Check which adhesive you pick up. I was using one which went on white but dried clear until it came time to pick up a new tube to adhere the CLEAR sheet. That tube was of course goes on white and dries white.

I put on a thick bead of adhesive around the frame so that when the sheet goes on it smooths out and makes everything airtight and watertight. This step would go much easier if you had someone to help you drop on the sheet.

I had used a handful of wood screws with washers to tighten down the sheet on the one side. It kept everything flat and even although the screw heads were a bit too high. Next time I'll use screws with more flush heads.

Because I used the wrong adhesive, I used the silver tape to cover the edges and mask the white. It also had a benefit of covering some foam edges and keeping direct sunlight off of the adhesive. I guess it wasn't such a bad mistake after all.

Step 10: As Buster Poindexter Once Said, Hot Hot Hot!


  • 4" Fan
  • 12v DC Adapter
  • Temp Probe

Ok, it's all together, now what?

Day 1:

For the first test I wasn't able to carry the unit out of the "Build Pit of Awesomeness" until later in the day. It was about six in the evening when I leaned the unit up against the garage and plugged in a small fan.

I used a 4" box fan salvaged from an old computer that was matted with a round PVC reducer. Fit perfect into the 4" tube. I hooked it up to a 12v adapter and dropped in the temp probe.

Wow, it got hot fast. The temp jumped up to around 120 degrees and then topped out around 135. The outside temp was around 87 and there were a smattering of light clouds. Not too shabby...

Day 2:

At least this time I was able to start a bit earlier in the day. The temp was also around 87 degrees but it was a bit past 2pm. Clouds were a bit heaver but the Sun was shinning. Again, WOW! The temp jumped once again but this time it topped out just a bit above 170 degrees. I think the heat sinkidea might have been a good idea because although the temp dropped a bit when a cloud floated past it was slow to drop and quick to recover. Ok, maybe I'm just reading a bit much into that last part.


Either way, the temp differential was pretty good and I'm hoping that during a cold Winter's day the extra insulation in the box will keep the differential just as high or higher.. We'll see how it goes when it gets cold. I've got a roll of 4" insulated tubing ready to connect this to an unheated space.

Update: I've swapped out the 12v PC fan for a 120v inline 4" ducted fan. Moves a LOT more air, very quiet and doesn't use up a lot of juice. If I can find a comparable DC version I may swap back because I'd prefer to use solar cells (without an inverter) to power the unit.


Step 11: How Does It Actually Work?

The stars aligned, the weather cooperated and I found some free time on my schedule so I decided to pop my heater into place and see if I could get it up and running. It's been, so far, a rather mild winter but I really wanted to get this up and running and see how well it works. Which is also, coincidentally, the time where all the warts start to actually show. At least I'm getting good data for my second build!! ;-)

Before I get into the, "what I would do better next time", let me start by saying that this thing PUMPS OUT THE HEAT! With a lack-luster fan (ok, a gas powered leaf blower on low) it still managed to maintain a healthy 130F degree output temp on a sunny and clear 30F degree afternoon. Although the poly was warm to the touch the box itself stayed cool so I wasn't loosing a lot of heat through that. The tubing was also rather cool. Big test comes with colder weather and a more sustained air throughput.


My plan was to always pipe the air in through a basement window but I wasn't sure how I was going to cut through a glass window. I was "prompted" to find a solution when one of my windows broke (don't ask) so I built a new wooden frame with a polycarb sheet and cut holes for my two flanges. Fit like a glove. One problem down.


I picked up some insulated 4" flexible duct at my LHS and secured that to all of my flanges. It consists of an internal spring tube, surrounded by fiberglass insulation and covered in a protective sleeve. Should be waterproof and should "help" keep the heat flowing. My biggest fear here is that the ends will become water logged. I used some silver flashing tape to wrap the ends and then zip-tied everything tight but water WILL get in the ends so I'm not exactly sure what I'll for a permanent solution. Suggestions always welcome!


During my initial testing I used a 4", 100 CFM, duct fan and it worked very well. What a fool I was. Since then I've learned a bit more and I now know why that fan isn't suited to blow out a candle. Static pressure is the force incurred when air is moving through a system. The longer the ducts and the more twists and turns in the air's way the more "push" is required to get the air moving. This little duct fan just couldn't make it up the mountain with all the added duct work. That impeller fan has a large gap between the tube walls which allows the majority of airflow to blow back through that gap. Although a leaf blower works GREAT as a test I decided to pick up a VenTech Inline fan. Handles the static pressure with ease.


Damper. Need a damper! Since the main unit is up and above the basement a convection current is created at night that causes cold air to flow back out through the feed pipe. Since everything is currently in "manual" mode I'm just stuffing some fiberglass insulation in the tubes when not in use, seems to do a decent job. However, I'm really looking for something that will automatically seal off the airflow when not in use. (Oh yeah, that will also increase the static pressure).


I've got this one! I have an Arduino and a couple of temp sensors that I'll use to automatically turn the fan on and off as the box temp rises and falls. I've got the code "mostly" written and sensor soldered to the end of an RJ45 cable run up through the feed pipe. Shouldn't be an issue. It's going to wait until the box is several degrees higher than the basement temp and then start blowing and as long as it's warmer than the basement it will continue to blow. If for some reason it stops, it will wait again until the box is several degrees warmer before re-starting. Eventually I'd like to get that system hooked into a Raspberry Pi solution I'm currently using which incorporates mySQL and Python for pretty graphs and such.


Before I hooked up the final connections and turned on the (pitiful) fan, I had my heat box angled for the best exposure. There was some moisture inside the box that condensed on the inside of the poly. when I removed the temporary vent coverings a lot of steam came out. It was HOT inside. I'm not so worried about summer as this will be coming down but what happens if I don't need to run the fan? How much heat can this thing handle without rupturing? For now (as long as I get the automation up and running) it shouldn't be an issue but if it stays in manual mode or the inside temp gets too warm there's going to be a point that this things starts to melt. Temp solution, throw a cover over the box. Ugly but it works. Long term? I'm thinking that there needs to be some sort of external cut-out that runs air through the box and just vents it outside to keep it from melting down.


Had the box in the garage for a bit and it attracted some dust on the outside. FILTER THE AIR THAT GOES INTO IT!! If you push air through the box without a filter it will start sticking to the inside of the poly. This beta build doesn't have a removable top so there's no way to get in there to clean it out. And yes, it WILL attract every floating piece of dust.


Just a reminder really. If you're planning on putting a box up somewhere make sure that you check for full sun throughout the entire day. I had my spot picked out and was very happy. Southern exposure, close to basement window, etc.. Found that there's a tree that blocks "some" of the sun late in the day. Not a show stopper by any means and given all the variables I'd still plop it down in the same location. Just keep it in mind if you're putting one up. Full sun is always best but even in the late afternoon, with some shadows cast by a now leafless tree, my box was still putting out heat.

Step 12: UPDATED: Feb 2017

Hey Folks! It's been nothing but cloudy for the past several weeks so it's been a bit tough testing everything out but I believe we've already found a few interesting items.

AUTOMATION: I've just about finished my Arduino temperature control system. Definitely a MUST for a system like this unless your sitting all day by the on/off switch. A "mostly" cloudy day can produce an hour of full sun which is enough to churn out some heat. Without the automation it's all lost. I'll post an Instructable about the Aurdino controls when I'm done.

SCREEN vs CAN METHOD: I liked the idea of the screen method allowing most of the air to come in contact with the solar absorption material (screen) where the can collector as all of that dead air between the cans and the glass. Perhaps the can method isn't so bad after all? I'm wondering how much heat is lost because the only thing between the hot air collector and the cold outside air is a single peace of plexi. I'd love to have two, side by side, to figure it out. Perhaps the next build will have a more expensive insulated glass if I go with an enhanced screen collector.

SCREEN SURFACE AREA: I've seen designs that use a single, flat, angled piece of screen so I figured I'd run a bunch of hills and valleys to collect more heat. If the box is pointed directly, at all times, at the sun then more peaks and valleys equals more surface area to absorb the heat. However, if the box is static mounted than too many peaks and valleys will block the sun from hitting only the tops of the folds.

HOW MUCH HEAT DOES THIS REALLY PUT OUT: Here's where the rubber meets the road. I had a truly sunny day so I had the blower running from 9:30am to around 4pm. Outside temp was in the 30's. Based on a PI box I had monitoring temp and humidity, the best I could do was warm things up about 2 degrees. My basement usually runs in the mid 50's during Winter with about 40% humidity. I ran a few numbers based on the temp increase and amount of air that had to be warmed. My calculations popped out 2,400 BTUs for the day. It's a big basement at 14,000 ft3. I believe those numbers are correct but I'm not 100%. Length, width, height and temp increase over time.

GLUE IS NOT THE WAY TO GO: I used construction adhesive to secure my Plexiglas cover to the box. Came out one day to find the top corner (hottest) had pulled off and was letting air out (and water in). I secured everything with screws and washers. Next go around I'll use a LOT less glue and make sure I have mechanical fasteners wherever possible.

Outside Contest 2016

Participated in the
Outside Contest 2016