Building a Homebrew Solar Hot Water Heater

Well this year I've set my sites on building a solar hot water heater to reduce my propane bills and
fossil fuel usage.
I am going to model the design off of the website's $1000 Solar Hot Water Heater Design.  It is a drain-back system with a collector and hot water storage tank.  I am going to try to build it for around $500 while maintaining the same functionality.

The goals of the project are as follows:

1.  Build it for less than $500 cost.

2.  Provide equivalent performance and capabilities as $1000 Solar Water Heater

3.  Design my own Differential Temperature Controller ( myDTC ) with Temperature Logging to verify/record operational performance.

(8/1/10: This is done now. You can buy the finished version below for $79 that comes with 8 channels of temperature logging capability )

If you are interested in a low cost differential temperature controller for you DIY project:

4.  Use recycled parts and equipment as much as possible.

Solar Collector Design


Well I started work on the solar collector portion of the design this weekend.  My daughter's friend had just replaced their windows and had the old ones leaning against their fence when I happened to see them and asked if I could have some of them.  I ended up getting 6 windows, each 24" x 48" casement type, for free.  Here are some of them leaning against the fence.

Next step was getting the nice window out of the frames.  This was really pretty easy.  Pound off the sill, open the window all the way so its past 90 degrees open, then take a screw driver and remove the metal straps and brackets holding it in the frame.  That takes about 10 minutes for each window.  Hereis a picture of the first window removed from its frame.

Being the salvager I am, I took off all the hardware and saved the wood screws.  There were lots of nice metal
straps, angle brackets, latches and screws from the job.  Here's my collection after all the windows were done.

Now I've got to build these windows into frames and add the solar collector material and PEX tubing to the back.
I have 48 sq ft of glazing to work with, which I think I will build into two 6x4 collectors.

Updated 3/14/10

Worked on building 1st collector.  First step was building a simple frame for three windows out of 2x4s.  Here's the frame
with a couple windows on checking for fit.

Next I mounted the windows to the frame with some screws.  I pre-drilled the windows to make sure I didn't split the wood.  I just put a screw in each corner so it would be easy to take apart if necessary.

Next I caulked all the joints, removes some loose paint and put a coat of Kilz sealer/primer paint on he box and called
that done until final painting and adding some final flashing around the windows.

Next I started working on the back panel.  1st thing was too cut some slats to create the runners that will be used to hold the pex in place on the backboard.  I bought some 1x4 and ripped them on the table saw to make 1x2s and the cut them to length for the runners.  Here are the runners all cut and predrilled for screw mounting.

Next I tried to figure out how to run the 1/2" PEX tubing on the board and make maximum use of my cut flashing pieces
that are 10"x11".  I installed the bottom runner on the board first, making sure I gave it some slope to ensure it would
have no trouble draining back when the pump is off.  I gave it a 1/2" of slope for each run of pex.  Here is a picture
of installing the 1st runner.

I noticed the runner had some bend in it and ending up fixing the ends in place then installing the middle screws
after aligning the runner to my straight metal ruler.

After installing the bottom runner, I installed the upper half of the channel by placing the pex tubing in between the two
and making sure they were snug together before screwing down.  I am figuring I will be able to press the PEX and the
thin aluminum sheets in this channel.  Here's a picture of putting in the top part of a channel.

Here is a picture after all the runners are in place and I have laid in the PEX tubing into the channel to see how its going
to work.

Last thing I did was check how my aluminum flashing pieces were going to fit and realized I needed a few more since I
ended up adding the fifth run of PEX.  Here is a picture of the small pieces of flashing laying on the board.  The rest of
this will need to wait till I get the rest of my flashing sheared.

Note: 8/15/10 -- After this was done and I was doing some testing on it I realized it would be better to have both inlet and outlet come out on the same side to reduce the outside exposure to ambient temperature and reduce losses.  FYI

3/22/10 - Update

Next thing was to preform each piece of aluminum with a simple form.  Here is the form, a piece of the flashing, a piece of the PEX and a 2x4 to press the PEX and flashing into the form.

Here are some of the preformed pieces on the back panel.

Since the flashing is so thin, it didn't hold its form.  I decided to mount it as follows.  Lay silicon caulk down the furrow where the PEX will be pressed on a row. Then press the PEX down into the form and use some large staples to hold the PEX down tight in the furrow. I had to do one piece at a time and kinda stand/kneel on my 2x4 foot press while nailing in the large staples.  I just put them in tight enough to just deform the PEX a tiny bit.

Here is the all the rows stapled in. 

Notice the fins of each piece are sticking up a bit.  From here I just put in some
small staple gun staples in the center of each piece into the furring strips to bring them flat.  Here is a picture of
everything stapled down.  The last thing I did was use some metal tape to cover all the PEX and provide some
heat transfer to the top of the PEX tubing. 

I added a small thermistor I built for sensing the temperature on the panel.

From there I just spray painted it flat black and screwed it into my window frame to give it a quick test.  Here it is
in the backyard.  Notice all the trees in the glass reflection.  I don't have a really sunny spot all day to test the unit.  I end up with shading for most of the afternoon.  This will be reduced when I trim back some trees.

Unfortunately, there was only a few 5 minute periods of sunny on the most cloudy day.  I managed to get it up to about 145F in about 5 minutes before it got cloudy.  Will run this on my diffferential temperature controller next weekend and take some thermal recordings of units performance.  One thing I noticed is that some of the staples that were holding the flashing flat pulled loose when it got hot, so I will need to add a few more staples to secure the flashing.  I also need to add some insulation to the back and sides. 

Lastly it seems like if it get set on or near the ground, a white / reflective footing might help increase the incident light on the collector.

Well, that's it for now.


I ran it for a day connected to my homemade temperature controller.  Here's a picture of the temperature controller I built to test it.  The temperature controller performs the differential temperature control needed for solar hot water collecting as well as records temperature data for 4 temperature sensors I built to go with it out of NTC thermistors and ring lugs.

The temperature probes on T3 and T4 were unconnected and have bogus readings.  Temperature Sensor T1 was the inside of the collector and T2 was the outside ambient.  Managed to get up to around 180F with no insulation on the back or even caulked up tight on the back panel.  Looks like it works good enough to try running some water thru it with the pump and storage tank.  Here is some capture from the run.  I got a dip later in the afternoon due to trees shadowing it, so I'm not sure how hot it could have gotten in the mid-day sun.

myDTC Deluxe v1.2
s)Status  c)Config  t)Set LogTick  f)Freeze Mode
0)Set SI Units  1)Mode_R1  2)Mode_R2  3)SP_R1
4)SP_R2  5)Bump Minutes  6)Bump Hours
7)DeltaT  8)LowT  9)HighT  m)Menu  r)Rules
 01 F   05 F   60 F  DTC_R1  DTC_R2  5 MIN    25 F   25 F  ALLOFF
Logging Tick :5 MIN
(u)Up (d)Down (e)Set (m)Menu
(u)Up (d)Down (e)Set (m)Menu

Time          Tick  T1  T2  T3  T4  R1  R2  R1TIME R2TIME
000:09:30  666  66  59 212  24 ON  ON  002.4   014.8
000:09:31  667  66  59 212  24 ON  ON  002.4   014.8
000:09:32  668  68  59 212  24 ON  ON  002.4   014.8
000:09:33  669  68  59 212  24 ON  ON  002.4   014.8
000:09:34  670  68  59 212  24 ON  ON  002.4   014.9
000:09:35  671  66  59 212  24 ON  ON  002.4   014.9
000:10:02  698  95  60 212  24 ON  ON  002.9   015.3
000:10:03  699  97  60 212  24 ON  ON  002.9   015.3
000:10:04  700  99  60 212  24 ON  ON  002.9   015.4
000:10:05  701 100  60 212  24 ON  ON  002.9   015.4
000:10:06  702 100  62 212  24 ON  ON  003.0   015.4
000:10:07  703 102  62 212  24 ON  ON  003.0   015.4
000:10:08  704 102  62 212  24 ON  ON  003.0   015.4
000:10:46  742 138  66 212  24 ON  ON  003.6   016.1
000:10:47  743 138  66 212  24 ON  ON  003.6   016.1
000:10:48  744 140  66 212  24 ON  ON  003.7   016.1
000:10:49  745 142  66 212  24 ON  ON  003.7   016.1
000:10:50  746 142  66 212  24 ON  ON  003.7   016.1
000:10:51  747 144  66 212  24 ON  ON  003.7   016.1
000:10:52  748 144  66 212  24 ON  ON  003.7   016.2
000:11:45  801 174  71 212  24 ON  ON  004.6   017.0
000:11:46  802 176  69 212  24 ON  ON  004.6   017.1
000:11:47  803 176  71 212  24 ON  ON  004.6   017.1
000:11:48  804 176  71 212  24 ON  ON  004.7   017.1
000:11:49  805 176  71 212  24 ON  ON  004.7   017.1
000:11:50  806 180  71 212  24 ON  ON  004.7   017.1
000:11:51  807 180  71 212  24 ON  ON  004.7   017.1
000:11:52  808 176  71 212  24 ON  ON  004.7   017.2
000:11:53  809 180  71 212  24 ON  ON  004.7   017.2 -- 180F inside temp at 11:53 with outside 71F.
000:11:54  810 176  71 212  24 ON  ON  004.8   017.2
000:11:55  811 176  71 212  24 ON  ON  004.8   017.2
000:11:56  812 176  69 212  24 ON  ON  004.8   017.2
000:11:57  813 176  71 212  24 ON  ON  004.8   017.2
000:11:58  814 172  69 212  24 ON  ON  004.8   017.3
000:12:20  836 163  71 212  24 ON  ON  005.2   017.6
000:12:21  837 163  71 212  24 ON  ON  005.2   017.6
000:12:22  838 160  71 212  24 ON  ON  005.2   017.7
000:12:23  839 160  71 212  24 ON  ON  005.2   017.7
000:12:24  840 160  71 212  24 ON  ON  005.3   017.7

I am really happy with the temperature sensor performance and the differential temperature controller operation.  Will know better once I have a pump to run with it and can get some collector efficiency numbers and measure the amount of heat I'm collecting in the tank and the tank losses.

Cost Info for 2 of these Solar Collectors:

Here is a workup of the cost of this solar collector (includes costs to build two of them).

48 SqFt Solar Collector Cost

48 SqFt Solar Collector Cost

Next step is too build up a solar storage tank and get the pump so I can run some testing to see how well the collector transfers the heat to the water.


Performance Testing the 24 SqFt Collector

I did some testing of this collector with a 5 Gallon tank to see how many BTUs it was
putting out.  Here is the test data.  The 1st sheet is calculations of how many BTUs
I should be available for water heating after collector losses.  The 2nd sheet is test
data for two separate tests I ran. I was able to get 2250 BTU/Hr out of this collector during
the test.  I think its possible to bring this up to around 2500 BTU/Hr by adding some more
insulation to the 2x6s inside and insulating and minimizing the PEX tubing length outside
the collector.

Thermal Calculations for Solar Collector and Tank

Thermal Calculations for Solar Collector and Tank

Here are some pictures from testing.  Basically the test setup consisted of:

1.  24 Sqft Solar Collector
2. 5 gallon bucket filled with water and insulated with some R-13 insulation
3. Submersible pump and some PEX tubing to pump water from the bucket, thru the collector,
and back into the bucket.
4. Some temperature sensors and my temperature controller to record the collector temps and water temps.
5.  All tests were run for one hour and BTUs and water heating recorded.  This data is recorded in the
spreadsheet above.

The first test I ran, the collector didn't have any insulation on the back of it.  I added it after trying to figure
out why the output was low.  Adding the insulation helped a great deal. 

Here's a picture of the pump I used for the testing.  I ended up turning the flow down some due to some
bad PEX to PEX fittings that had a slow leak.  I figure I was running about 1/2 rate, maybe 165gph.

Here is the collector and 5 gallon water tank with insulation under the work bench.  The pump is inside the
bucket.  A computer above is recording temperature via my temperature controller. The return is on the other

Here is a picture from the second days test were I've added some insulation to the back of the collector.  In the
picture above there is no back insulation.  You can also see just a piece of the PEX return tubing that goes back to
the tank.

Hot Water Storage Tank Design


I decided to abandon the original enclosure idea in the hot water storage tank page because I became worried that the loose insulation would get wet due to the high humidity of the hot water.  The other problem is the other enclosure got really big due to the 6 inch walls. A buddy at work had some 3 inch styrofoam he wanted to get rid of that he was taking out of an old hunting cabin.  So I got some of it and decided to make my enclosure out of it. 

Here is a picture of a piece of it as I'm getting ready to cut some wall pieces out.  It cut easily with a kitchen knife.

Here are 4 walls cut and you can see the base piece under the barrel.

Next I got the jig saw out and cut out a big hole in the top of the barrel to get the heat exchanger and pump in and out easily.  I drilled 4 corner holes with a 1/2 inch drill and then used a jig saw to cut it out.

I decided to wash out the barrel good and took it outside to clean it up.  At this point I needed to wire up my water pump to the myDTC temperature controller.  I am going to use the controller to turn the pump on when the solar collector is hotter than the water in the tank.  I am using 3 or the 4 temperature inputs on the myDTC:

T1 - This is measuring the solar collector panel temperature
T2 - This is measuring the temperature of the water in my storage tank.
T3 - This is measuring the ambient temperature

Here is a picture of the temperature controller and a AC plug wired into the temperature controller relay output.  The water pump is plugged into the plug which will get power when the solar collector temp is hotter than the water.

Here's a link to my website I use to sell differential temperature controllers with logging capabilityIt's got 8 temperature probe inputs, and works for differential, heat and cool mode thermostat, provides logging of temperature data to flash and costs $69.

Here is the temperature sensor for the water.  I took the sensor and put a big piece of aluminum tape on it and hung the sensor in the barrel until the aluminum tape was submerged.  Eventually I will waterproof the temperature sensor, but for now this will work...

Here is a picture of barrel showing the water pump at the bottom of the barrel.
  I am planning on using a swiftech DC pump in the final version, but this submersible AC pump is good for now.

I lashed the styrofoam walls together with some rope for now to do some outside testing.  Eventually the tank will be inside the laundry room next to the hot water heater.  The enclosure isn't perfectly air sealed since I want to be able to take it back inside and its too big to fit thru the door all put together.

In the picture you can see the PEX tubing coming in and out of the tank and the lashed together tank.  I ended up cutting some little tunnels at the top edge to fit the PEX under the cover piece.

I put the top on and started the temperature controller and programmed it to log temperatures every 15 minutes.  I will check back in a couple of days and provide the logging data so we can see how well it worked. 

Here is some data for 8/15 - 8/16 - It was sunny for some small stretches, but mostly cloudy with some rain showers.  The yellow line is ambient temperature, Blue is collector temperature, Red is the Hot Water Storage Tank Water.  For two kinda cloudy days, it actually performed pretty good.   The tank enclosure is still not sealed so its probably leaking heat a little more than normal since I can see some big cracks at the seams.

Temp Testing 8_14_15

Solar Hot Water System Block Diagram

Drainback Block Diagram

Basic operation of the system is as follows:  Solar energy heats the solar collector plates, increasing their temperature.
Once the temperature is greater than the temperature of the storage tank water but not above an over temperature setting,
the pump is activated, pumping storage tank water thru the solar collector and back to the storage tank, thereby
increasing its temperature.  When the solar collector temperature cools below the water tank temperature, the pump is stopped and the solar collector water drains back into the tank. 

The collected heat in the storage tank is transferred to the cold inlet water to the hot water heater by cycling it thru a heat exchanger coil in the hot water storage tank which heats the inlet water to the hot water heater. This reduces or eliminates the need for the hot water heater to heat the inlet water.

If you have any questions you can add a comment or email me at:

Thanks for visiting,

Brian Smith

Copyright 2010 by Brian Smith

The material on this website authored by me, Brian Smith, is copyrighted.

You must have permission from me to use any of the text or pictures on these pages.