A standalone dual-pass solar air heater using downspouts

  • Active since 1995, Hearth.com is THE place on the internet for free information and advice about wood stoves, pellet stoves and other energy saving equipment.

    We strive to provide opinions, articles, discussions and history related to Hearth Products and in a more general sense, energy issues.

    We promote the EFFICIENT, RESPONSIBLE, CLEAN and SAFE use of all fuels, whether renewable or fossil.
Status
Not open for further replies.

precaud

Minister of Fire
Jan 20, 2006
2,307
Sunny New Mexico
www.linearz.com
As part of my ongoing project to reduce wood consumption, I've just installed another solar air heater, designed and built incorporating some of the concepts that I learned from woodstoves over the years. It's a simple design that uses aluminum downspout tubes as the absorber and heat exchanger. That in itself is nothing new; downspout tubes have been used in solar heaters for a few years and have proven to be a good choice for horizontally-displaced air heaters. But the way this one works addresses the major shortcomings of earlier designs.

Previous tube designs were basically insulated boxes with tubes attached to the back and glazing on the front. Air flows through half of the tubes in one direction, then reverses through the other half. And therein lies the two major problems. Attaching the tubes directly to the back, separated only by an R6 insulating board, creates a heatsink directly from the heat exchangers to the outdoors, cooling them off. And the long path length that the air travels through is not an efficient use of the exchanger surface area, requiring high air velocity through the system to achieve reasonable efficiency.

Both problems can be solved with minimum added complexity by suspending the tubes away from the back, using that channel to rout the incoming air to the other end, and then flow it through all of the tubes at once in parallel. The heat that was previously conducted through the back now preheats the incoming air, and air velocity is cut in half or better.

All of the materials needed to build this can be found at your local big box store, total cost around $200.

This is what the finished heater looks like, vented into a basement window to heat my lab (yes, the duct tubes will be insulated at some point...). The dimensions are 12' L x 2' H x 8" D. The heat exchanger area is 10' x 2', which will heat 200-400 sq ft. depending on the insulation, height, etc. I'm guessing it will reduce my wood use downstairs by a third or more.

A snap switch is built in for fan control, turning on at 110F and off at 90F, making it completely automatic.

Today is the first day I'll be able to operate and test it - should be fun!
 

Attachments

  • bhtr15.jpg
    bhtr15.jpg
    75 KB · Views: 2,535
Very cool. I'd like to understand your airflow better. Have any more pictures that would help illustrate?
 
Sure. The first pic shows the unit face down, the tubes are placed on the front panels to position them.

The second one shows the baffle, which isolates the input and output air.

The third one shows one of two 3/16" steel rods that support the downspouts. And everything is siliconed in place. It is important to have no gaps anywhere, including between the tubes.
 

Attachments

  • bhtr2.jpg
    bhtr2.jpg
    72.1 KB · Views: 2,681
  • bhtr3.jpg
    bhtr3.jpg
    71.2 KB · Views: 3,325
  • bhtr4.jpg
    bhtr4.jpg
    60.2 KB · Views: 2,071
The first pic shows the first section of the back in place, showing the 2-1/2" gap between the back and the tubes, where intake air is preheated.

The second pic shows the section of the back that mates with the input/output baffles.

The third pic shows the baffles with the front cover removed so they can be sealed up. The snap switch is wired in at this point.
 

Attachments

  • bhtr7.jpg
    bhtr7.jpg
    55 KB · Views: 1,997
  • bhtr8.jpg
    bhtr8.jpg
    59.7 KB · Views: 1,932
  • bhtr9.jpg
    bhtr9.jpg
    59.7 KB · Views: 1,762
The first pic shows the front of the unit with tubes installed.

The second pic shows the back with duct connectors and painting completed (yes, I built it in my living room... there are advantages to not being married!)

The third pic shows the front with snap switch and glazing supports.
 

Attachments

  • bhtr5.jpg
    bhtr5.jpg
    65.2 KB · Views: 1,619
  • bhtr10.jpg
    bhtr10.jpg
    34.1 KB · Views: 1,580
  • bhtr11.jpg
    bhtr11.jpg
    55.6 KB · Views: 1,559
The first pic here shows the glazing is in place.

Second pic shows the ducting into the basement window, which will be insulated soon.

The third pic shows the 7" fan roughed in. The fan is on a variac so the optimum air flow throughput can be found. The intake duct will extend down to near the floor and mate up with an input filter and backdraft damper.
 

Attachments

  • bhtr12.jpg
    bhtr12.jpg
    68.4 KB · Views: 1,460
  • bhtr13.jpg
    bhtr13.jpg
    86 KB · Views: 1,462
  • bhtr14.jpg
    bhtr14.jpg
    45.5 KB · Views: 1,413
And lastly we have the output temp in degrees F measured by a Fluke temp probe at the end of the output duct. 65F in, 120F out at this flow rate. I like it.
 

Attachments

  • bhtr16.jpg
    bhtr16.jpg
    54.6 KB · Views: 1,088
Neat. I love your projects. Is the ducting just PVC pipe? Have you considered insulating the pipe to reduce heat loss?
 
BG, yes, the ducting is PVC sewer pipe. It will be insulated shortly, I have the material, just waiting for a little warmer day to do it.
 
This is awesome Precaud. Great work, and great presentation as well.
 
Thanks Dune. It's working well, and will work even better when I get the ducts insulated over the next few days. I just got my old Anemotherm anemometer working, so soon I'll be able to present credible output btu/hr data. And once I get the solar radiation input variables figured out, I can calculate overall efficiency. Fun stuff.

Added: The output duct is now insulated, it's just R6 foil-faced fiberglass and not a permanent solution, but it made a nice difference.
 

Attachments

  • bhtr17.jpg
    bhtr17.jpg
    122.8 KB · Views: 989
I ran measurements on the collector today. It was clear all day except the last hour or so, when some high clouds moved in. See the attached graph of output temps, from 8:40am to 4pm.

With solar radiation data calculated for this latitude and the collector's orientation, a collector area of 18.5 ft^2, airflow of 65.4cfm, and an air density correction factor of 0.83 for the 7,000 ft. altitude, we get this set of numbers for btu/hr output and efficiency over the burn cycle I mean collection cycle:

Local time_____Tdelta______btu/hr_______Efficiency
8:50am________35.92______1950________45.4%
9:50am________44.45______2412________45.3%
10:50am_______51.32______2785________48.6%
11:50am_______50.32______2731________48.6%
12:50pm_______47.62______2585________50.1%
1:50pm________41.42______2248________54.2%
2:50pm________29.96______1626________55.2%
3:50pm________15.71_______852________53.5%

Without the correction for altitude, the efficiency would be a good 10% higher across the board. Sigh. But even in its present form, it is giving me 17,200 btus per day for the cost of running a 20 watt fan for 7 hours. A pretty good deal.

I'm told that the best commercial collectors are in the 50% efficiency range, so this is comparable to slightly better. I could probably eek a bit more out of it if I insulated the intake duct. That will have to wait until next year - it's supposed to start getting cold here in the next few days, so its time to get the woodstoves ready.
 

Attachments

  • Oct22.gif
    Oct22.gif
    34.1 KB · Views: 840
Not too shabby precaud. Free heat is free heat. I will be curious to see if the numbers drop mid-winter when intake heat loss becomes more of an issue, or not.
 
Yes BG, I'm pretty pleased with it. Would have liked to have made it twice as big, but there wasn't room for anything bigger. I too am interested to see the numbers when it is truly cold out. My bigger heaters aren't affected by it, but they don't have exposed input ducts, either. Plus, if there's snow on the ground, efficiency actually improves from the reflected light.

karl, flow rate is air speed (measured at the intake) times duct area.
 
Nice work. I have never seen downspouts used. I love the fact that you are harnessing the free heat and putting 120 air into your basement. Very cool... er warm. I am doubling the size of my basement window heater and will just need glazing to finish. I'll have more pics up on that thread soon. I love free heat. I just need sunny days.
 
I have another question. If you used a smaller or lower rpm fan, could you realize a higher output temp as the air would be in the collector longer, or is overheating the materials a concern?
 
fishingpol said:
I have another question. If you used a smaller or lower rpm fan, could you realize a higher output temp as the air would be in the collector longer, or is overheating the materials a concern?

He's getting on average a 40 degree delta. That's pretty good. Plus 120 is hot. Without going to evacuated tubes, I don't see how you could get it much hotter in the winter.

I bet a bigger fan would result in more total btus extracted.
 
fishingpol, I look forward to seeing your collector details.

Someone in another forum suggested I try using Reflectix Reflective Foil Insulation on the intake duct. So I went and got a roll of it this morning. I must admit I've been skeptical about this stuff in the past, but as we'll see, it appears to work pretty well in this application.

The roll I got is 16" wide. The duct pipe is 4-1/8" OD, putting the circumference approx. 13.5". I have some 1/4" heavy-duty paper-backed bubble wrap laying around, so I put a layer of it on the pipe first to build it out. That worked out very well, adding another insulation layer and making the diameter perfect for the 16" Reflectix.

After sealing it up with Gorilla Tape and giving the system a half hour to adjust, I logged temps every hour from 11:50am to 3:50pm to compare with yesterday's data. The results were better than I expected; temps up 3F to 5F across the board, with approx. 200 btus more output per hour:

Local time_____Tdelta______btu/hr_______Efficiency
11:50am_______53.77______2919________51.8%
12:50pm_______51.12______2774________54.7%
1:50pm________44.38______2409________58.1%
2:50pm________34.10______1851________62.9%
3:50pm________20.05______1088________68.3%

So BG, your intuition was correct: a 6' long bare intake duct was costing about 1400 btus per day, roughly 9% of total output.

Lesson learned: Make ducts as short as possible, and be absolutely FANATICAL about insulating and sealing everything in the air path! (Sound familiar? I've been saying the same thing about air/combustion systems in wood stoves...)

I won't be doing any more measurements on this thing until it turns much colder out. The hunt for quieter fans is all that remains to be done for now.
 
karl, you're right. The sweet spot for output temps appears to be in the 110-120F range for best efficiency. With the fan on a variac, I've run tests at 45 cfm, temps peaking at about 135F, and efficiency was about 5% lower. At 65.4 cfm, it tops out around 123F. Something around 70-75 would be ideal. Pulling that with 0.5" H2O static pressure with low noise levels in a 4" deep space is the next puzzle to solve.

btuser: That’s about $.40 worth of heat/day.

Measured in what? And what's your point?
 
Couple of more questions:

I am curious why the snapstat is in the collector housing? It just seems like a bit of work to get at it if it fails. Is this where the first usable temps are realized to start the fan?

Could this design work if say air flowed horizontally down 2 spouts and then did a 180 degree turn and came back on the next 2 above it and so on? If the air flowed longer through the collector back and forth, wouldn't it pick up more temp along the way?

_____

Sorry Karl, I am not really a formula and numbers guy. I just run ideas through my head and hope for a good number in the end. Building these are half the fun without getting to complicated. I just think of any way to extract more juice of the the giant lemon called the sun.
 
fishingpol said:
I am curious why the snapstat is in the collector housing? It just seems like a bit of work to get at it if it fails. Is this where the first usable temps are realized to start the fan?

Where else would you put it? Yes, it will be a pain to get to if it fails. A temp sensor with leads back to a controller in the house would have been more elegant.

Could this design work if say air flowed horizontally down 2 spouts and then did a 180 degree turn and came back on the next 2 above it and so on? If the air flowed longer through the collector back and forth, wouldn't it pick up more temp along the way?

Nothing of heating value happens in a 180 degree turn. It only adds back pressure to the system, requiring a more powerful fan to maintain cfm throughput.

IMO, these downspout heaters are only a good choice for heaters with horizontal form factors. It's better if you can go vertical, take advantage of natural thermosiphoning, and use black screen for the collector/heat exchanger. Cheaper and easier to build, too.
 
Ok, thanks for the answers. I am used to looking at thermosiphons, so I need to wrap my head around the horizontal set ups to understand them a little more.
 
Glad to see insulating your output helped. I would suggest insulating the input. There won't be as big of an improvement as the input, but it's still warmer than ambient temperature.

I'm impressed hat you're getting over 900 btus a day per square foot. I wonder if the soda can collectors would do as well as the downspouts. I suppose they would, and be much cheaper.
 
Status
Not open for further replies.