Norm's system uses a large homemade solar heat storage and drain back tank to provide solar space heating and domestic water heating for his home in Maine. The system integrates the solar heating with a boiler that provides the space and water heating functions when the solar water is not hot enough.
The tank, plumbing and heat exchangers are very nicely done and provide a lot of good ideas and construction details.
Its also a good example of getting professional help with some parts of the system while doing the rest yourself.
Thanks very much to Norm for sending this in!
Norm's full description...
I acted as my own general contractor and subbed out the pieces to people with considerable experience in each. The overall design of the heat distribution system was done by Northeast Radiant Technologies of Gardiner, Maine. One facet of their business caters to DIYers. They also made suggestions on boiler specs that were passed on to George Phelps of Gilman Plumbing and Heating of Newport, Maine. George worked with Quincy Hydronic Technology of Portsmouth, New Hampshire. They are the distributors of Biasi boilers in New England. George's son Rob installed the boiler and all the related plumbing and controls. The solar portion of the system was designed and speced by Vaughan Woodruff of Yankee Solutions of Pittsfield, Maine. Vaughan worked with both both Gilman and NRT to ensure that all parts of the system were compatible and complementary. I built the tank and coils and did all the solar plumbing and control installation. Vaughan and I installed the solar collectors. The overall system underwent many changes from the original concept to the final installed version. I will not go into these here, but would be willing to discuss with anyone that might be interested.
Installed system with boiler at upper left.
This system combines a 550 gallon drainback solar tank combined with a conventional propane fired boiler (Biasi Riva Plus). The solar tank has two 60' copper coils to heat domestic hot water and two 90' coils for space heating. Either part of the system can operate independently of the other or they can work together. A temperature sensor in the tank is attached to a relay that cuts power to the boiler any time that the tank temperature is greater than 125 F. This temperature was arrived at by running a flow test and determining that the DHW coil in the solar tank will deliver 120 F water continuously when the tank temperature is 125 F. Also the space heating coils will deliver essentially 125 F water to the input of the boiler. Since this is above the normal temperature required for supply to the floor, there is no need for the boiler to run. Both sets of coils in the solar tank can be shut off with ball valves, if need be, and then the boiler will provide DHW and space heating. The input and output of both sets of coils and the supply and return of the radiant floor have inline thermometers to provide information on system operation.
There are four SunEarth Empire EC-32 collectors, attached to the south wall of our log home, that have water pumped through them by a three speed Grundfos Alpha pump. I have the pump set to operate at the middle speed. The pump is controlled by a Solar Thermal differential controller that senses the water temperature at the top and bottom of the tank and also the temperature of the air inside the collectors. The controller is set to allow a maximum temperature of 140 F inside the tank. If the tank temperature is below 140, the controller is further set to start the pump any time the collector temperature is more that 16 degrees above the bottom water temperature. The controller turns the pump off if the collector temperature falls to within 8 degrees of the tank water temperature. At this point all the water in the collectors drains back to the tank. The controller has a safety feature which prevents starting the pump if the collector temperature is above a preset upper limit to prevent steam generation. The circulation piping has a Pentair flowmeter which shows water flow rate while the pump is running and acts a sight glass for tank water level when the pump is idle.
Click on pictures for full size
The four Sun Earth collectors
This system is used to provide heated water to two different types of radiant floor heating systems. One part of the heating system is the conventional Pex tubing in concrete floor in the basement area of the new addition. The first floor of the addition uses Warmboard with Pex-Al-Pex under ceramic tile. The water is circulated through the boiler and the tank heat exchangers by a variable speed Grundfos pump. The individual zones are regulated by thermostatically controlled valves. At the present time there are four zones working and plans are underway to install four more loops in the old part of the house. These will replace the existing forced hot air system currently in use.
Warmboard radiant floor heating.
(Click on schematic for full size)
System schematic diagram
As shown in the schematic, all items in green are parts of the space heating
side of the system and the items in red are part of the domestic hot water side.
In normal operation valves 2,3,5 and 6 are normally open to allow water to flow through the respective heat exchangers. Valves 1 and 4 are normally closed. If for any reason, it is desired to have the boiler operate without any boost from the solar tank, valves 1 and 4 would be opened and 2,3,5 and 6 would be closed.
Pump A is part of the boiler installation and pump B is dedicated solely to pumping water to the solar collectors.
Heat exchanger 1 consists of two 90' coils of 3/4 copper connected in parallel. Heat exchanger 2 has two 60' coils but otherwise is the same as 1.
The boiler that I used has a primary heat exchanger for space heating and a secondary for DHW.
The tank was sized to accept a liner purchased from Tom Gocze at American Solartechnics in Searsport, Maine. The inside dimensions are 55x55x48 h. The tank is made from four panels for the side walls, a platform and a removable cover. The cover is sealed on the edges with 0.5 round foam insulation to cut down on evaporation loss. The construction of the tankverticallpanels, platform and cover are described below. panels, platform and cover are described below.
The 550 gallon solar heat storage and drainback tank
(click on pictures for full size)
The two exposed vertical panels had another 1.5 of rigid insulation applied to the outside and a final cover layer of OSB to protect the insulation and create a mounting surface for the boiler piping and controls. The corner that the tank was installed in has insulated foundation blocks for the first 18 of height and a very well insulated wall (R 45)above that. There is also a floor drain in the corner and the floor is pitched so any leaks should not flood the basement ( barring catastrophic failure). The drain is also used for the condensate drain from the boiler.
Tank structure diagram.
(click on diagram for full size)
The liner that I purchased resembles a water bed mattress with an open top. I did not install the liner until I had put the coil support frame and coils in the tank first to be sure that everything was going to fit without any contact with the side walls. I did not want to risk puncturing the liner during this process. I then removed the coil support frame and the coils and installed the liner. I made sure there were no sharp edges inside the tank and vacuumed the inside to remove any unwanted debris. I then placed the liner in the tank and got inside with my shoes off and smoothed out wrinkles and worked air bubbles out from between the liner and the tank. The liner has a flap that laps over the top edge of the tank panels and these were secured temporarily to hold the liner in place. At this point, I put on a bathing suit and started slowly filling the tank while making sure that the liner remained in contact with the tank along the edges and in the corners. I filled the tank to a depth of about 18 and then installed the top rim of the tank to hold the liner in place. This top rim also has holes to accept the 1/2 copper tubes that support the coil support frame. I then installed the coil support frame and the heat exchanger coils.
Nearly finished tank.
A part of the tank liner can be seen on the back inside wall of the tank.
The tank panels are a sandwich of two sheets of 5/8 OSB glued and screwed to a framework of 1.5 thick strips ripped from 2x4 and 2x6s. One sheet of OSB was applied to the frame, the voids filled with 1.5 rigid polyisocyanurate and then the other sheet of OSB was applied. The horizontal strips inside these panels were placed closer together near the bottom, since this is where the max load occurs. The actual width of the strips, except on the vertical edges is not really critical to the overall strength of the panels. I used lag bolts to fasten the panels together at the corners and thus needed to have sufficient material for these to grab into. I also used rafter plates that I bent into ells to further solidify the corners. These panels and corner connections are no doubt overdesigned but I had most of the material on hand and the the time needed to precisely calculate the optimum dimensions of the components couldn't be justified. I did do some basic strength calculations to ensure that the design was at least adequate.
Tank sidewall panel before insulation and inner OSB face sheet were installed.
The tank platform is constructed similar to the tank side panels. The major difference is that 2 insulation was used in the platform instead of 1.5. The platform is supported on pressure treated 2x4s to get the OSB up off the floor. The space between the PT 2x4s has 1 rigid insulation glued in to create a final insulation thickness of 3 under the tank, the same as the finished vertical panels. The platform was created larger than the horizontal dimension of the vertical panels to allow for the extra insulation and OSB.
This picture shows the tank before the outer insulation and OSB were applied. The vertical panels are in place on the platform. The entire perimeter of the the vertical panels is captured by a strip of OSB that is glued and screwed to the top surface of the tank platform. This provides extra blowout protection ( probably not necessary) and ensures that the vertical panels retain their proper alignment with the platform (very necessary) while the tank is maneuvered into its final position. The corners are secured with 5 galvanized lag bolts that go through the face of one panel into the end of its mating panel. The head of the bolt is visible just below the corner anchor plate. The anchor plates are secured with stainless screws in both legs.
This picture shows the coils in the tank without the liner in place. This was part of a test run and I wanted to be sure that the liner did not get damaged. The coils having the wider spacing are for DHW. The right hand pipe enters the tank above where the liner will eventually fold over the top edge of the side panels, goes to a point mid way between the coils and goes to a tee at the bottom of the coils. The water then circulates up through the coils and exits through the left hand pipe. The other set of coils operates identically. The only differences are the overall length of the coils and the change in spacing to accommodate the extra coils. The coils are supported by a pipe frame to keep them in the top half of the tank. The coils were assembled in pairs outside the tank and then hoisted into the tank using a lightweight block and tackle. In order for this to be possible, the coils were fitted with stiffeners to stabilize them. These stiffeners are shown closeup just below.
Heat exchanger coils positioned in tank.
In both HX the incoming water is introduced to the coils thru a Tee that connects the bottoms of the two coils. Therefore 1/2 of the water being pumped goes thru each coil. The water then comes back together thru a Tee at the top of the HX and continues on. The space heating HX1 has two 90 coils of 3/4 copper formed in a coil with a 22 ID. The DHW HX is the same except the coils are only 60 long.
With a water temp of 140 F in the tank water being pumped thru the HX by the boiler circulator exits at nearly 140. Thermodynamically, I realize this is impossible. The only explanation that I have is that the tank temp is measured by a probe in the tank that sends a signal to the solar pump controller and the exit water temp is being measured by an inline analog thermometer. I am sure that the exit temp is lower than the tank water temp but my guess is that it is within one to two degrees and the thermometers are not sensitive enough to tell exactly.
I have tested the DHW HX by running the hot water in my kitchen sink for 15 to 20 minutes and checking the temps at the end of the run. Again, with 140 F water in the tank, the inlet water is usually about 50 F and the exit water temperature is about 135 F. I have a probe in the tank that is connected to a relay that controls the power supply to the boiler. This relay shuts off power to the boiler anytime the tank temp is above 125F. This provides 120 F water for DHW and approximately 125 F water for radiant heating. Our radiant system is designed for water temps below 120 so the boiler doesnt really need to run when the tank temp is above 125 F.
To form the coils, I created a form using the wooden ends of a large wire spool. The center of the spool was removed and the ends were cut with a jigsaw to have the same OD as the desired ID of the coils to be formed. The ends were then attached together with short pieces of 2x2 to complete the form. Be sure to make the 2x2's long enough to accommodate the number of coils that you will be making. I screwed the form to my bench, anchored one end of the coil of tubing and using a rubber mallet worked the coil to reduce the diameter to match the diameter of the form. This process requires a little patience but actually goes quite quickly.
Form for making the heat exchanger coils.
Heat exchanger coil on the form.
This picture shows the support frame in the tank without the liner. This frame was constructed from a combination of 1 and 1/2 rigid copper pipe. The horizontal portion of the was sized to keep it and the coils away from the tank liner to prevent any chafing of the liner. Its purpose is to support the heat exchanger coils in the upper portion of the tank where the water should be warmer. The return pipe from the collectors is part of this frame and supports one corner. The rest of the frame is supported by 1/2 pipe risers that are capped and inserted into holes in the top rim of the tank walls, one in each corner. This top rim will be above the liner once it is installed. Even though this frame will eventually have holes to allow the water out, it was initially assembled to be watertight. This frame was pressure tested with air to be sure that there were no leaks in the risers.
Heat exchanger coil support frame.
In the picture , the vertical pipe at the far right is the suction pipe for sending water out to the collectors. The other vertical pipe close to the suction is the return pipe from the collectors. This return pipe has a 1/8 hole drilled in it just below the elbow. This serves as a vacuum breaker to allow the drainback to work properly. The vertical pipe on the left is used to support one corner of the frame. There are two more of these pipes in the other two corners. The horizontal pipes in the corner of the frame furthest from the suction have several 1/4 holes drilled in the bottom of them to let the return water out. This ensures that the heated water returning from the collectors is inserted back into the tank as far as possible from where the water to the collectors is being drawn out. I drilled enough holes so that the cross-sectional area of the holes was slightly larger than the area of the return pipe. The holes should be far enough apart to not compromise the structural integrity of the pipe.
The coil stiffeners/spacers shown were created by heating and compressing short lengths of 1-1/2 PVC drain pipe. Once a coil spacing was decided on, the required length of pipe was cut and then heated using an electric heat gun. A propane torch could also be used but it is easy to scorch the pipe. Once heated, the pipe is very pliable and can be flattened using a piece of 2x4 and some clamps on the edge of a bench( or two pieces of 2x4 and clamps. Be careful to leave one edge with an eye to accept the zip tie which holds the pipe in the stiffener. Let the pipe cool back to room temperature and it becomes rigid but in a new shape. It can then be drilled and cut to create the slots for the pipe. When putting these on the pipe, I found it easier to start at the top and insert the zip tie as each loop of the coil was inserted. I created 12 of these and put three on each coil. This changed the coils from unwieldy 60' or 90' slinkies into something manageable. I created a slide rail above the tank, and using a light block and tackle was able to lift two connected coils into the tank. In addition to making the coils manageable, these pieces also hold the coil loops apart to aid in the heat transfer process.
Heat exchanger coil stiffeners.
The system as shown has been in operation since April of 2011 and has performed very well. We had been heating hot water with electricity and our electric bill has gone down $25 to $35 per month. The overall performance on the heating side is tougher to gauge due to many changes related to the installation of this system in conjunction with building a very well insulated addition to our existing house. The addition is on the north side of a log home. The R value of the original logs is estimated to be about 7.0. The walls of the addition have an R value of 45.0. The new construction added about 30% to the overall floor area and I estimate the heating cost has been reduced about 15%.
June 4, 2012
blog comments powered by Disqus