First Try At An Over and Around the Window Collector

The Mother Earth News folks have a strong interest in a solar air heating collector that works in conjunction with a regular window.  This is based on the long term popularity of the "Heat Grabber" collector that MEN published a number of years back.  While the Heat Grabber has been popular and works well, it does suffer a bit from limited collection area and, because it sticks out from the building making it difficult to use in  some situations.  So, the MEN folks are interested in anyone's ideas on how to improve the design.

over window collector

This is my rough first prototype effort at a solar air heating collector that could be installed over a regular window.  The idea is to use the existing window to get air into and out of the collector.  This avoids any need to cut holes in the wall for vents.  It also makes it feasible to just install the collector over the winter.  My main objective was to make the collector area large enough to make it really worthwhile to put the work into building the collector.

So, there is plenty of room for improvement in all aspects of the collector -- how about some ideas?

Some things to keep in mind:

- The window itself is a good solar collector, so the part of the collector that covers the actual window is not really very helpful at all in collecting heat -- the window itself would do about as well.   So, in this over window concept, the collector has to be significantly larger than the window it is covering.
- For solar heating, area counts a lot, so the larger the collector can be made the more heat it will collect, and the more worthwhile the work invested in the project will be.
- Stagnation temperatures can be fairly high (see below), so the materials used must be able to stand some exposure to fairly high temperatures -- perhaps up into the 200F area for a single glazed version.
- The vertical orientation selected for this prototype appears to have some advantages: 1) it does not stick out from the wall,  2) it may allow the collector to blend into the house and look better (with some trim work), 3) it allows for a larger collector area, and 4) vertical collectors benefit from the reflection from any snow surface in front of the collector.
- Air heating collectors work much better with absorbers that have a large surface area, and that have a good distribution of air flow over the full area of the absorber. 

The material below is a first cut at trying to do this over window collector, and I'd appreciate any ideas anyone might have on how to improve it.  In some ways, it has some of the same challenges as  David's Flow Organizer, even though it does use a fan.


If you have any ideas that you think might work for improving this design, or you have an alternative, please send them in to me, or, you can send them into Mother Earth News.



This was put together in a big hurry as we are packing for a trip that starts tomorrow morning.

The  Concept


First Test




The Concept

The prototype surrounds an existing window that is about 33 inches by 50 inches, or 12 sqft.


The prototype is 69 inches wide by 94 inches high -- 45 sqft -- so it does provide quite a bit more collection area than the existing window.


Both supply and return are located at the bottom of the existing window.  You open the window about 7 inches, and insert the supply/return gadget.  My windows only open at the bottom, so this is about the only way to do it -- if your windows open top and bottom, you could use the top opening for the return air.


At the top of the collector, there is a 4 inch diameter hot air collection manifold.  The collected hot air goes down a 3 inch diameter duct from the center of the manifold down to the slot at the bottom of the window, the 3 inch diameter duct takes a right angle turn into the room, where a fan mounted on the end of the duct pulls air through the duct and expels it to the room.  So, its a fan forced collector.


The air supply vents for the collector are also located in the slot at the bottom of the window.  There is one on each side, and each is 3 inches by 10.5 inches.  The vents extend into the collector about 4 inches.  The hope here was that the the cool supply air would follow the cool glazing downward -- as it does on with David's  Flow Organizer.  If this works, it eliminates the need for some kind of ducting system to take the incoming air tot he bottom of the collector.


The absorber is black insect screen (as used on the barn collector)  -- I like this kind of absorber for the window collector because it lets you look out the window, and it also looks better from the street side, as you can see the siding through the screen.   Window screen can make a good collector absorber...


Note that I was in a rush to get this done, and did not have anything good to use for glazing.  I ended up piecing together two scraps of vinyl -- this is why the tape splice across the glazing.



The shutters make this look dark, but if there were siding, it looks fairly good through

the absorber.




One window with collector, and identical window without.  Note that the collector could be made wider to add more collector area.






this shows the hot air manifold at top with many holes in its bottom surface.  The 3 inch

hot air duct goes straight down the from the center of manifold, and then turns into the bottom of

window opening.  The fan is just the other side of the brown board.

The two inlet ducts with the sheet metal extensions are on either side of the hot air duct.

Idea of the sheet metal extensions is to encourage the incoming air to get over to the glazing a

and sink down the glazing, and then be heated and rise up the absorber.

The screen absorber is positioned such that most of the inlet vent air has to go through the absorber to get out.

The screen is two layers near the center of collector, and one layer outside -- this is from just overlapping the 48 inch material I had.



From a seasonal installation point of view, everything in the picture above can be a single piece that is hung on a couple lag screws under manifold, or by a couple "log screws" through the manifold.




This is the inside view.

This could, of course, be made to look nicer, and the fan need not extend so far into the room.

This is a window in my shop/garage, so it has no inside trim or finished sill.

The vertical height of the brown board could also be less.








This is the collector outer frame that supports the glazing.

For the prototype, is just rigid foam board insulation glued together with Great Stuff.

The Vinyl glazing is mounted in a wood frame that sits on and is glued to the foam board.

A final version of the collector would probably want to be made from a more durable material.





This picture shows the wood frame for the glazing on the foam board.



Another view of the finished collector and frame.



So, just to recap the concept.

Air enters one of the two inlet vents, and is directed toward the glazing.


Idea is that this air will make its way over to the glazing, and because it is cool, and the glazing is even cooler, it will sink down the glazing (as in the Flow Organizer).


As air flows down the glazing or when it gets to the bottom of the collector, it gets heated by contact with the absorber and begins to flow up the absorber and through the absorber. 


The two inlet vents are as wide as possible, and located as far outward as possible to encourage flow across the whole width of the collector.

Trying to avoid dead air spots with poor circulation.


As the air flows up the absorber it must flow through the absorber in order to get to the top hot air collection manifold.  The "flow through" absorber is intended to improve heat transfer from absorber to air.


Both the pressure gradient created by the fan, and the buoyancy of the heated air bring it up to the hot air manifold, where it is whisked away down the 3 inch duct, through the fan, and into the room.

At least, that's how its supposed to work.


First Test

I just had time for one quick test today before we leave tomorrow.


This window in on the west side of the house, the sun comes directly onto it around 4pm local time.


During the test, the sun was a few degrees south of the window in azimuth with an elevation of 33 degrees.  It appeared to be quite clear.  In other words, good collecting conditions.


I ran the fan for about an hour, and collected a few inlet and outlet temperatures.  Also had a go at measuring airflow (less successful), and measured stagnation temperature.


Also had a go at a smoke test for flow visualization.


Representative temperatures:

Tinlet = 78F

Thot air outlet = 118F


So, a 40 F rise. 


The stagnation test resulted in 160F in 15 minutes of stagnation -- it might have been increasing very slowly when I stopped.


Ambient temperature was about 75F -- possibly a bit higher right at the wall.



Stagnation test sensor placement.

Flow volume.

This little Sunon 12 volt fan is just the little fan that could.  Qualitatively, it appears to put out a lot of air.


I tried to measure flow rate three ways -- none completely satisfying:


The fan outlet velocity.

Using the Kestrel wind meter, the peak of the velocity profile was just over 1600 fpm.

If I assume the average is 1500 fpm, and associate a diameter of 3 inches with the flow, then the flow rate comes out about 120cfm, or 2.65 cfm pr sqft of collector -- a lot.

These assumptions on average velocity and flow area are subject to a lot potential error.

Flow velocity in inlet vents.

I measured the flow velocity in the inlet vents using the Dwyer Vaneometer.  There was lots of variation, and it appeared that this was due to the wind pushing the Vinyl glazing in and out.  At most, the velocities were around 180 fpm, and at least near nothing.


If one were to use 100 fpm as an average, the flow rate for the collector would be about 50 cfm.


One factor here is that the collector frame is not sealed to the house, so there could be a fair bit of exterior air being pulled in the gaps around the frame.


Garbage bag flow rate

I had a go at trying to do flow rate by timing the fill time for a 45 gallon plastic garbage bag.

This is the first time I have tried this, and it would have benefitted with more time to get it set up right, but we were out of time.

As nearly as we could jury rig it, the fillings appeared to take about 6 seconds.  This works out to about 60 cfm.


So, hard to say exactly what the flow rate was, but maybe somewhere between 60 and 100 cfm.  It does look like its up in the 1.5 cfm per sqft, which seems good to me.



I think the whole thing would benefit from a 4 inch duct and a bigger, slower, quieter 4 inch fan.


So, going out on a limb, the heat output at this stage appears to be around


(90 cfm)(0.07 lb/ft^3)(118F - 78F) (0.24 BTU/lb-F) = 61 BTU/min, or 3700 BTU/hr  -- 82 BTU per hour per sqft.


While the 3700 BTU per hour is worthwhile, the 82 BTU/hr-sf does not indicate a killer efficiency -- clearly there is work to do  -- see smoke test results below.

Smoke Test:

I had a go at a little smoke test.

I put a smoke "pill" in the north inlet duct, and lit it.


See the picture below

On the plus side, the smoke appears to cover north half of the collector pretty well in the right to left sense.


On the negative side, the scheme to try to get the flow to go down the glazing is a total failure.  It appears that all of the smoke goes up.

So, the bottom third of the collector below the window might as well not be there.


One other observation is that the collector and fan do a dandy job of filling the shop with smelly smoke.






Any comments or ideas on how to improve the collector from either a performance, looks, construction or life point of view are very welcome.



On the performance side (which was the main reason for doing the prototype),

Is there a way to encourage entering air to follow the glazing down and make better use of the bottom of the collector?

This seems to work well on the Flow Organizer, but is a dismal failure here -- what am I doing wrong?

Maybe the inlet vents need to go through openings in the absorber and terminate closer to the glazing?

maybe this scheme does not work well until the weather is cold enough to cool the glazing down more?

Maybe the pressure gradient created by the fan overpowers the rest?


Or, should I just go to extending the inlet ducts down into the lower area, at the cost of more parts and build complexity?


Any other ideas on how to get good airflow over the full collector with both supply and return at the bottom of the window?


It would be nice if there was a way to make the collector work without a fan at all, but I don't see how?


On the construction and aesthetics side:

For a collector of this size, the foam board construction seems doubtful.  Not sure what would be a better frame material choice.

Something that looks more finished might also help.


The glazing clearly needs to be better -- maybe a thin, rigid, flat polycarbonate film?  Maybe even twinwall polycarbonate?


While the stagnation temp was fairly low, I think that when the frame is sealed to the wall, it will go up, and the PVC ducts probably need to be metal.


Any other suggestions?


It does seem to me that if a house had 2 or 3 windows that this could be applied to, you could get up toward 100 sf of collector, and with some efficiency improvements, this could provide some very useful heat for house.

Comments -- Suggestions

blog comments powered by Disqus





Gary September 11, 2009

Updated April 11, 2010