musingsheader image
0 Helpful?

Solar Hot Water

Heating water with the sun isn’t cheap

Posted on Aug 28 2009 by Martin Holladay, GBA Advisor

If you’re aiming to reduce your carbon footprintAmount of carbon dioxide and other greenhouse gases that a person, community, industry, or other entity contributes to the atmosphere through energy use, transportation, and other means. , you’ve probably thought about installing a solar hot water system. Here’s the good news: if you have an unshaded south-facing roof, you can install a solar hot water system that will meet about half your annual hot water needs.

The bad news: the typical solar hot water system costs between $6,000 and $10,000.

ICSIntegral collector storage solar water heater., Drainback, or Antifreeze?
There are three main types of solar hot water systems (also known as solar thermal systems):

  • Integrated collector-storage systems;
  • Drainback systems; and
  • Antifreeze systems.

Integrated collector-storage (ICS) systems, also known as “batch” systems, store hot water in a roof-mounted tank above the collector. Such systems are common in Israel, Hawaii, and other warm climates where freeze damage is unlikely.

Those of us who live in frosty climates need a freeze-resistant system. To protect against freeze damage, a solar hot water system either drains all the fluid from the collector when it’s cold — the drainback approach — or circulates an antifreeze solution through the collectors.

The collectors in a drainback system are usually dry. When temperature sensors indicate that the sun is shining, a control activates a pump which circulates water through the collector pipes. Later, when the control senses that the sun has gone away, all of the water in the collectors is automatically drained to a “drainback tank” — a special tank, separate from the home’s hot-water storage tank, that holds the collector water. The next day, when the sun comes out again, the water from the drainback tank is again circulated through the collectors. Whenever the pump is operating, hot water from the collectors circulates through a heat exchangerDevice that transfers heat from one material or medium to another. An air-to-air heat exchanger, or heat-recovery ventilator, transfers heat from one airstream to another. A copper-pipe heat exchanger in a solar water-heater tank transfers heat from the heat-transfer fluid circulating through a solar collector to the potable water in the storage tank. to raise the temperature of the water in the main storage tank.

Propylene Glycol Antifreeze
The second way to make a solar hot water system freeze-resistant is to use an antifreeze solution in the collector loop. When the sun shines, a pump circulates the antifreeze solution; after being warmed in the solar collectors, the fluid passes through a heat exchanger and gives up some of its heat to the water in the hot-water storage tank.

Regardless of which method of freeze-protection you choose, you’ll have to make two other important equipment decisions:

  • Do you want flat-plate collectors or evacuated-tube collectors?
  • Do you want an AC pump or a DC pump?

Copper Plates or Evacuated Glass Cylinders?
Flat-plate collectors are glazed, insulated boxes containing copper tubing mounted on black-painted copper absorber plates.

Evacuated-tube collectors consist of an assembly of glass cylinders, each enclosing a partial vacuum. When the sun shines, metal elements inside the evacuated tubes get hot; the heat is conducted to a fluid circulating in a manifold at the top of the collector.

Flat-Plate Collectors Are Simple and Robust
I’m a fan of flat-plate collectors:

  • The technology is simple and time-tested.
  • They cost less than evacuated-tube collectors.
  • On sunny days, they perform better than evacuated-tube collectors.
  • On an annual basis, they provide more energy per square foot of collector and more energy per dollar invested than do evacuated-tube collectors.
  • Snow melts faster from flat-plate collectors than from evacuated-tube collectors.

Evacuated-tube fans like to point out that flat-plate collectors experience more heat loss at cold ambient temperatures than evacuated-tube collectors. (True enough — if the snow ever slides off the evacuated tubes). Moreover, evacuated-tube collectors perform better during cloudy weather, and begin collecting heat earlier in the day, than flat-plate collectors.

Remember, though, that you’re not going to be collecting much useful energy on cloudy days or very cold days anyway. As explained by the author of a technical bulletin from EnerWorks, a Canadian manufacturer of solar equipment, “On overcast days the evacuated-tube collector will perform better than a flat-plate collector. Of course, if there’s not much sun to begin with, doubling your efficiency is not a big advantage. The question is, do you want a collector that will perform better when there is plenty of sun to be captured, or one that will perform better when there is not much sun to start with?”

PVPhotovoltaics. Generation of electricity directly from sunlight. A photovoltaic (PV) cell has no moving parts; electrons are energized by sunlight and result in current flow.-Powered Pumps Are More Dependable
There are two types of pumps used in solar hot water systems: AC pumps and DC pumps. AC pumps use line-voltage electricity; they are usually activated by a controller responding to temperature sensors. DC pumps, on the other hand, are usually wired directly to a small photovoltaic(PV) Generation of electricity directly from sunlight. A photovoltaic cell has no moving parts; electrons are energized by sunlight and result in current flow. (PV) module.

You should choose a DC pump — ideally, the El Sid pump from Ivan Labs in Jupiter, Florida — powered by a 10-watt or 20-watt PV module. Such a pump will be much more dependable than an AC pump activated by a differential-temperature controller and sensors — especially since AC pumps don’t operate during a power failure.

Don’t Forget the Downside
Let’s say you’re a fan of solar power, perhaps because you’re old enough to remember when Jimmy Carter installed solar collectors on the White House roof. (Ronald Reagan did not appreciate the symbolism of a solar-powered White House; he had the collectors taken down soon after he became President.)

You’re ready to shell out $9,000 for your own solar hot water system. (Since you’re probably eligible for a 30% Federal tax credit, your actual out-of-pocket expenses will only be $6,300.) You’d like to have two solar collectors, a 120-gallon solar storage tank, and a lot of insulated pipes filled with a glycol solution.

Before you sign the contract, remember:

  • Glycol antifreeze solutions degrade over time. These fluids are quickly damaged by “stagnation” during sunny weather; stagnation occurs when a pump fails. That’s why pump dependability is so important.
  • As part of normal maintenance, it’s a good idea to check the pH of your antifreeze solution every year or two. If the solution has become acid, it’s time to replace the antifreeze.
  • When it’s time to re-roof your house, your solar collectors will have to be temporarily removed at significant expense.
  • Unless you live in Hawaii, don’t expect your solar hot water system to save you any money any time soon. A 2006 study by Steven Winter Associates calculated a 58-year payback period for a system installed in Massachusetts, while a Wisconsin system had a payback period of 76 years. The longer payback period in Wisconsin was due in large part to the system’s AC pump, a significant parasitic energy load. Because the Massachusetts system had a PV-powered pump, it had a quicker payback.
  • The more expensive the energy you use for your conventional water heater, the sunnier your climate, and the higher the available rebates or tax credits, the faster your payback will be. Investing in a solar hot water system makes much more sense for an Arizona homeowner with an existing electric water heater than for a Minnesotan with a natural-gas water heater.

I Installed One Anyway
Three years ago, ignoring the bleak payback calculations, I installed a solar hot water system on my roof. My system includes two 4' by 8' collectors, an 80-gallon stainless-steel tank with an integral heat exchanger (although the tank was affordable, it’s smaller than ideal), an El Sid pump connected to a 20-watt PV module, and an expansion tank. (The hot water in the storage tank is connected to a stainless-steel coil in my woodstove; the water in the wood-stove loop circulates by means of a passive thermosyphon.)

Here’s what I learned:

  • The best book on designing and installing a solar hot water system is Solar Hot Water Systems by Tom Lane.
  • If you’re willing to do all the work yourself, you can install such a system for about $3,000 or $4,000.
  • Solar collectors are big and awkward — much bigger and more awkward than PV modules.
  • It’s important to insulate your pipes well. Protect all rooftop pipe insulation with foil-faced butyl flashing tape.
  • Be meticulous when soldering and tightening threaded joints. You don’t want any leaks.
  • For me, the most challenging tasks were pressurizing and bleeding the air from the antifreeze loop.
  • It’s fun to watch the temperature gauges rise on a sunny day.

Conservation Is Always More Important Than Energy Production
Before concluding that you need to produce more energy, it’s always best to try to lower your energy consumption. Be sure that you have an efficient washing machine, a low-flow showerheadShowerhead that restricts water flow to less than the 2.5 gpm limit (at 80 psi) mandated by the U.S. EPA., an efficient plumbing layout, and a conserving lifestyle. All of these steps will yield a better return than a solar hot water system.

Last week’s blog: “Can Foam Insulation Be Too Thick?”

Tags: , , , , , ,

Image Credits:

  1. Robb Aldrich, Steven Winter Associates

Aug 28, 2009 8:34 AM ET

The web site
by Fred

The web site is a great resource for DIY solar projects, including a solar water heater the author built for around $1,000.

Aug 28, 2009 8:42 AM ET

Some caveats
by Martin Holladay, GBA Advisor

I admire do-it-yourself projects. But anyone attempting to build the system described on the "builditsolar" Web site should know:

- Many people have experimented with large storage tanks built of EPDM and plywood. While some of these are successful, many others have been plagued by leaks. You don't want a huge water tank in your basement to begin leaking.

- The DIY system uses several plastic components, including PEX and black polyethylene pipe, in the homemade solar thermal system. These components may fail at high temperatures.

There's nothing wrong with backyard tinkering. But most people will find that commercial collectors and commercial water tanks are less hassle and much more long-lived than homemade components.

Aug 28, 2009 9:07 AM ET

by Jason Miller


Do you have any knowledge of or experience with the Hot2o system by Fafco? ( It was featured in Fine Homebuilding back in 2007, and I've kept it in the back of my mind ever since.

I like many aspects of their system, but I haven't actually seen it in operation or talked to anyone who has used it or installed it.


Aug 28, 2009 9:29 AM ET

Fafco plastic collector
by Martin Holladay, GBA Advisor

I reviewed Fafco's Hot2O plastic solar collector for the May 2007 issue of Energy Design Update. Fafco promotes the plastic collectors as part of an inexpensive one-tank system. Here's an excerpt from the EDU review:

“[According to] Jay Burch, a senior scientist at the National Renewable Energy Laboratory, … 'If you have a one-tank system where the second [electric] heating element is not turned off, the system will perform poorly. Even if the thermostat is at low setting -- say, 105 degrees or so -- that’s still a pretty high starting point. If you can’t turn the lower element off, then you might as well just forget it. And if you do turn the lower element off, you take a hit in terms of available hot water when there is no solar availability. With a 40-gallon tank, you have about 35 gallons of more or less hot water. But if you turn off the lower element, you won’t have the volume of hot water sitting there, and on a cloudy series of days, you’ll only have 10 or 15 gallons of hot water. Consumers will have problems with that. Of course, if you have a gas water heater, you will always have to go to a two-tank system.'

"Other possible disadvantages to the Fafco Hot2o system:
- According to Fafco, the plastic collectors are expected to last only 12 to 15 years.
- Water circulating through an unglazed plastic collector doesn’t get as hot as water circulating through a conventional glazed collector. While glazed collectors can easily heat water to 190 or 200 degrees, “The Hot2o can heat water up to 150 degrees,” according to Doug Kohl, a representative from Fafco’s public relations firm.
- In some areas of the country, the limited heat that can be extracted from Hot2o’s collectors may not justify the cost of the electricity required to circulate water through the collectors."

Aug 28, 2009 7:50 PM ET

Nice recap
by Carl Seville, GBA Advisor

Thanks for the overview of Solar thermal. In case you aren't aware of it, Velux has a nice, simple flat plate system that attaches to the roof with their well established skylight flashing system, making them less likely to leak initially and eliminating the need to remove the collectors when reroofing.

Aug 28, 2009 8:45 PM ET

Solutions to removing collector for reroofing
by Mark Bartosik

Velux (yes the skylight manufacturer) has a collector that flashes under the roofing material like a skylight but without the hole in the roof deck. I am going to be fitting one in a few weeks. I think that it looks better too this way.

I'm considering having the collector frame spray painted to match the roofing tiles. The roofing tiles are a world first -- recycled TPO synthetic slates (like EcoStar - but by Authentic roof) but this is the first batch ever in a light color with infrared reflective pigment. Maybe I'll submit a blog entry for this.

There's the other solution to reroofing problem - use a roofing material that won't need replacing before the collector. The slates should last over 50 years.

Aug 29, 2009 4:14 AM ET

Velux collectors
by Martin Holladay, GBA Advisor

Carl and Mark,
It's interesting that you both mentioned the Velux collectors, which are being fairly heavily promoted and advertised these days. I haven't actually seen these — I've just seen the literature. I think the jury is still out on how these units will perform.

Tom Lane, the author of Solar Hot Water Systems, is a fun author to read because he's a cantankerous, opinionated curmudgeon who also knows his business. At the end of his book, he has a two-page chapter called "The Worst Solar Hot Water Ideas of the Last 25 Years." Among these bad ideas are collectors flashed to the roofing — in other words, Velux-style collectors. Lane calls these collectors "integrated roof collectors." (Kind of like building-integrated PV — he's talking about roof-integrated solar thermal collectors.) Here's what Lane writes: "Integrated Roof Collectors — A Service Nightmare. Difficult to weather-seal, often impossible or extremely difficult for service and maintenance. During stagnation temperatures, there is a possibility of charcoaling the roof trusses and setting the house on fire. [In contrast,] Collectors mounted on aluminum rails 1 1/2" off the roof will have a pleasing appearance without these issues. Integrated collectors cost more and can leak into the attic."

Aug 29, 2009 7:22 AM ET

Solar water heaters for an avg. family of 4 persons in 3rd World
by daud

I am interested to work out a system for an average family of 4 persons (to max 8 persons) living in Third World countries — Africa, Middle East, India, Afghanistan, Pakistan. The cost of 100-liter water heaters ranges from $125 to $150 (made in China). The remaining cost of installation varies.

I understand it is a simple hot water arrangement to bath and kitchen, and the temps range from -4 degrees C to 45 degrees C.

These areas have plenty of sun throughout the year.

Please guide me — What can be a system to provide heating throughout the year for these low-income people?

Aug 29, 2009 2:33 PM ET

Velux review
by Michael Chandler, GBA Advisor

We installed one of the Velux collectors in a house last year. It was an absolute dream to install, we cut BIG holes for the plumbing connections and once the collector was racked to the roof and flashed and shingled in we went below and made all the connections in the shade standing on staging we had set up in the roof trusses.

It's a propylene glycol system and here in NC I'm partial to drain-backs due to summer over-heating issues and the need to run the circulator at night (if the storage is over 170 at 2:00 am) to cool the storage tank and keep the system from steaming off when the family goes on vacation. But the install is so elegant that I'm tempted to use it for all domestic hot water only situations with no radiant heat assist implementation.

The collector intake and output are both located at the top of the panel where work space is more ergonomically convenient (a feature I've always appreciated with the tube collectors but which could create a serious problem with steaming off after a hurricane where you would have an extended clear sky power failure.) The connections from the control module to the panels is made with pre assembled corrugated flex hose with unions on both ends so there is no soldering required in the propylene glycol lines. Line sets are available at different lengths for different roof configurations. The storage tank is 80 gallon with a coil-in-tank heat exchanger.

I prefer a dual pump (drain-back) set-up with a flat plate heat exchanger like the Solar Hot ( but this is a very solid and durable solution for the average suburban application where looks, durability, ease of maintenance and speed of installation are more critical than lowest first cost and most BTUs per square foot of collector area.

Aug 29, 2009 2:52 PM ET

Another solution to vacation overheating
by Martin Holladay, GBA Advisor

Thanks for the very interesting review. Here's what I did to address vacation overheating: I installed two tees, above and below the vertical check valve that normally prevents nighttime thermosyphoning. The tees are connected to create a bypass around the check valve; there's a ball valve in the bypass. This is my "vacation bypass" valve. The ball valve is opened when I leave during the summer, to allow natural thermosyphoning to occur at night. The glycol in the loop is heated by water in the storage tank; cooled in the collectors; and in turn cools the stored water. No electricity needed.

The detail came from Tom Lane's book.

Aug 29, 2009 3:54 PM ET

Tom Lane solution
by Michael Chandler, GBA Advisor

Nice trick! I can see that this would work with a coil-in tank system, Maybe with a two pump flat plate system but I think the Velux detail where the intake and output both come out of the top of the collector would possibly create a thermal dam, still worth a try. I'll see what Dan Gretch has to say on this one.

Aug 30, 2009 11:43 AM ET

by Carl Seville, GBA Advisor

Martin - Interesting that Tom Lane calls integrated panels difficult to weather seal since the Velux panels use the same flashing systems that have worked incredibly well on skylights for many decades. I can't see how they could be more difficult to maintain than rack mounted systems, particularly since you can reroof around them without removing them, certainly much easier than rack mounted systems that have to be taken off the roof.

Also, his position that rack mounted collectors have a "pleasing appearance" is very subjective. Personally I prefer the look of integrated panels, although I appreciate his concerns about overheating the roof structure.

Aug 30, 2009 3:10 PM ET

Velux flashing
by Martin Holladay, GBA Advisor

To be fair to Tom Lane, his book was published before Velux began distributing their collectors in the U.S. Lane was apparently reacting to collectors that were "roof-integrated" by builders who installed improvised flashing systems and site-built details. I'm sure the Velux flashing system is a lot slicker than the details that influenced Lane's opinion.

Aug 31, 2009 3:29 PM ET

Thanks for the clarification
by Carl Seville, GBA Advisor

But it still doesn't mean I will be fair.

Aug 31, 2009 6:14 PM ET

Hi, I'm Gary from Build It
by Gary Reysa

I'm Gary from Build It Solar -- since my system was mentioned above, I thought I would add my 2 cents in.

Very nice summary of solar water heating systems.

On your comments on my system (the $1K Solar water heating system) I'll disagree a little on some of the points you made.
- All of the data I've been able to find on EPDM lined tanks that are limited to temperature below 170F has been good. There are several commercial makers of EPDM lined tanks, and they seem to have a good track record. I have heard from people who installed these tanks in the 80's and are still on the first liner! As far a leak potential, I think that the system really has less leak potential that conventional systems that use solar storage tanks that are pressurized. The non-pressurized tank in my system can leak out its full 160 gallons, but no more, since its not connected to the water supply. A conventional pressurized solar storage tank is connected to the water mains, and could leak thousands of gallons of water if you happen to be away.

- The system does use PEX tubing and also uses a pump that is only rated to 140F, but the differential controller is set to limit the tank temperature to a maximum of 140F. The PEX and the pump are very high quality components as long as the temperature limits are observed. These choices allow significant saving in system initial cost without any compromise on system life -- thats what good design is about.

I think the system goes a bit beyond backyard tinkering. I'm a retired product development engineer, and I put the same kind of care and testing into developing this system as I did products back a Boeing. You can have a look at all the testing work that was done leading up to this design here:
The goal was to do a system that was substantially less costly without compromises in materials, life, maintenance, or performance. And, to do a system that will perform year round in harsh climates with good performance.

I think that you really hit the nail on the head when you pointed out that solar water heating systems cost $6000 to $10000. This initial cost is so high that its very unlikely that these systems will be widely adopted -- most people simple won't spend $8000 to save a few hundred dollars a year.
My aim was to dramatically cut the initial cost without cutting the life, quality, or performance.
If I can do it, I see no reason why there can't be commercial product that achieves the same thing.


Aug 31, 2009 11:00 PM ET

Hi Gary
by Mark Bartosik

Gary: I'm a frequent reader of your site ( It is one of my favorites.
Yep costs are much higher for professionally installed packaged systems. I'm doing DIY. I've also got 3 old collectors that I plan to ground mount and use some ideas from your site - like do that as drain back to an unpressurized home built tank. That will be in addition to the Velux on the roof.

Martin: As for flashed collectors risk of charring the deck -- I'll check with Velux product manager (I've got his number). I have the option of installing 0.5" foil faced polyiso under mine since I'll like add 0.5" of ply to the surrounding deck (for other reasons).

If you look at my site: you will see that I am a fan of Building Integrated Solar. But I do acknowledge that the risks are higher -- you really don't want a wiring fault with BIPV or a leak with building integrated solar thermal.

But you have to admit my roof with the BIPV looks nice!

Aug 31, 2009 11:05 PM ET

Alternative stagnation solution
by Mark Bartosik

If the controller supports it (if not use a snap disc on the tank), then upon stagnation temperature alarm, operate a valve to dump hot water from the storage tank to a drain (or return of my geothermal).

Of course no problems with drain back. But I like the Velux collector which is not compatible with drain back, and I didn't want noise from drain back above my bedroom. That nice trickling sound would induce too many bathroom visits!

Sep 1, 2009 6:32 AM ET

An example of a leaking tank
by Martin Holladay, GBA Advisor

Thanks very much for your comments, and congratulations on building a solar thermal system that works for you.

One example of a leaking EPDM-lined tank was the first solar storage tank built by Eric Doub of Boulder, Colorado. (Doub is a GBA advisor and a pioneer builder of net-zero-energy homes.) I interviewed Doub about his experiences building a 6,000-gallon site-built tank in his basement. Here's what I wrote in the September 2006 issue of Energy Design Update:

"Doub’s biggest headache has been his storage tank. Worried about leaks, he decided at the last minute to change his original tank design (a concrete tank coated on the inside with a spray-on truck-bed liner). For increased waterproofness, he installed a layer of 40-mil EPDM rubber on top of the spray-on liner. Soon after the tank was filled with water, however, it began to leak, so Doub drained the tank and removed the EPDM. The useless truck-bead liner left a bumpy surface on the inside of the tank. To create a smoother surface, Doub installed a layer of ½-inch polyisocyanurate insulation on the tank’s walls and floor, covering the rigid foam with two layers of 60-mil EPDM. 'We spent about $10,000 for the tank,' said Doub. 'If I had to do it again, I would not install the type of tank we have in our basement. It’s a potential Poseidon Adventure.' ”

Doub now uses recycled stainless-steel dairy tanks (bulk milk tanks) for solar storage.

One question: with your system, what happens when your storage tank reaches 140°? You tell us that "the differential controller is set to limit the tank temperature to a maximum of 140°F," but you don't tell us where or how the heat is dumped.

My stainless-steel storage tank is able to handle water that is 180°F or 190°F. It's great to make use of extended sunny weather to increase the temperature of the water in my storage tank. I'd hate to have to begin dumping heat at 140°F.

Sep 1, 2009 9:03 AM ET

by Gary Reysa

Hi Martin,
On the tank, it sounds like in the initial installation they might not have taken enough care to prepare a smooth surface for the tank liner? You have to have a smooth surface, and its also very important to make certain that the epdm lies flat against the tank wall before adding water-- if there are areas where the epdm bridges across a tank corner, it will be stretched when the tank is filled. 6000 gallons is one very big tank! -- my domestic water heating tank is 160 gallons, and space heating tank is 500 gallons.

My controller has the max tank temperature switch set to 140F, so the pump is not turned on any more once the tank gets to 140F temperature. Since its a drain back system, all of the water in the collector and plumbing drain back to the storage tank at this point -- there is no fluid in the collector. So, the collector itself is the only part of the system that is exposed to the stagnation temperatures. My feeling is that collectors have to be designed for stagnation anyway, because at some point in the life of the system they will be exposed to stagnation temperatures. For most people this will occur whenever they go on vacation and there is no demand. Even if you have a heat dump, at some point in the life of the system there will be a failure that will expose the collector to stagnation.
Since there is no fluid in the collector, and the system is not closed or pressurized, there are no worries about boiling water in the collectors and blowing pressure/temperature relief valves and making big green puddles. This approach is pretty common in drain back systems.

My feeling is that if you have a somewhat more than the amount of storage used in most commercial systems, that there is no real benefit in going above 140F on storage. In the winter, when it would be nice to have water above 140F to cover the possibility of a week of stormy weather, the few hours of sun and low ambient temperatures keep you from getting the tank above 140 anyway. In the summer when there is lots of sun and warmer temps is easy to get the tank above 140F, but there is no real benefit in that there is so much sun that extra heat you store from 140 to 180 is not needed. At least, that has been our experience -- I've monitored the performance of the system for a full year now, and the solar fraction is above 95% with the 140F limit.


Sep 1, 2009 9:47 AM ET

Measuring solar fraction
by Martin Holladay, GBA Advisor

How do you measure solar fraction?

Do you have a flow meter on your domestic hot water supply pipe?

Have you logged the temperature of your domestic hot water supply?

How do you measure fuel use when your solar equipment provides 100°F water, and your backup system raises the temperature of that water to, say, 130°F?

How do you measure fuel used for backup? Have you separated out that fuel use from other usages using the same fuel in your house?

Very few homeowners have the monitoring equipment to accurately measure solar fraction.

Sep 1, 2009 5:04 PM ET

solar fraction
by Gary Reysa

Hi Martin,
I log the storage tank temperature.
I also log collector, tank bottom, solar radiation, supply and return to collector, but the only one I use for the solar fraction calculation is the tank temperature near the top.

I use an incoming cold water temperature of 50F -- I've measured this a number of times, and its always pretty close to this. Our water comes from a 240 ft deep well, and its temperature does not vary much.

Our target hot water temp is 110F. I know this is a little low for some people, but its what we use, and works fine for us. The dishwasher heats it above that for dish sanitation. I don't account for the heat the DW adds in the solar fraction calc, but I suspect most solar fraction calculations don't.

So, I do solar fraction as (Tstor - 50F)/(110F - 50F) .
So, to get 100% solar fraction for a day, the Tstor has to be at least 110F.
If Tstor is 100F, then the solar has heated the water from 50F up to 90F, or 40F out of the required 60F for a solar fraction of (90F-50F)/(110F - 50F) = 0.67 or 67%.
At the end of the month, I plot the tank temperature and do this solar fraction calc for the afternoon of each day (this is when our major hot water demands are) -- I average all the days, and thats the solar fraction for the month.

The heat exchanger on my system stores enough water within the heat exchanger itself at full tank temperature that there is no heat exchanger drop to account for -- the 12 gallons of water in the heat exchanger is already up to full tank temp. If the draw is more than 12 gallons (very very unusual for us), then the there is a small temp drop through the heat exchanger coil that is not accounted for.

Our backup heater is a small gas tankless heater. It only fires when the storage tank temp is less than 110F to bring it up to 110F. It is completely turned off most of the year, because its just not needed -- in these months, the solar tank temp never drops below 110F. I don't account for the fact that the tankless is only 85% efficient, but with a solar fraction over 90%, how much difference can that really make?

You can see the plots for nearly a full year here:

I don't account for the power used by the circulation pump, but its only 13 watts, so its really not going to make any significant difference (a couple hundred BTU per day).

Our hot water demands are modest (probably about 22 gallons a day), and we don't set our hot water target temp relatively (110F). We have a system that for the size of our demand has both more collector area and more storage than most commercial systems would be sized for. The tilt of the collectors is 70+ degrees and there is a reflective snow field in front of the collectors -- this optimizes winter collection, which is where most systems don't do so well. So, basically the system is oversized and optimized for good winter performance -- this is basically why our year round solar fraction is high.
With the cost of adding a few more square feet of collector area at $5 per sqft and another 50 or so gallons of storage at almost no cost, it seems to make sense to oversize the system compared to the "normal" ground rules?


Sep 1, 2009 6:00 PM ET

Your calculations are only approximate
by Martin Holladay, GBA Advisor

I admire your data collection, but your calculation of solar fraction are only approximate. You have pointed out a few issues yourself: for example, you fail to account for pump energy use (which Robb Aldrich's study has shown to be a significant factor in most solar thermal systems).

Moreover, you fail to account for actual gallons of hot water drawn; nor do you account for the temperature of the tank at the time the water is drawn. If five gallons of hot water are drawn at 7:00 a.m., then you assume that the temperature of the water in the solar storage tank is the same as in late afternoon, which is unlikely.

Moreover, you fail to account for the possibility that some large water draws may be made at a time when the temperature of your solar storage tank is less than its peak temperature, whereas the gallons drawn in late afternoon may (on some days) be few.

An accurate calculation would require a flow meter to measure the gallons of water used; temperature logging of hot water at the time of use; and logging of gas consumption used for your backup heater.

I certainly realize that such measurements are difficult — which is why there is a dearth of good data on solar fraction. I admire your calculations; but it's important to point out to readers that your solar fraction is far higher than published data from monitored systems, and this discrepancy is probably due to the fact that your calculations are only approximate.

Sep 1, 2009 8:21 PM ET

solar fraction
by Gary Reysa

Hi Martin,

The pump power is 13 watts (a very efficient pump). If it runs 6 hours on a typical day, thats (13*6) watt hours, or 266 BTU. During that time the system collects about (48 sf)(250 BTU/sf)(0.5 efic)(6 hours) = 36000 BTU, so not accounting for the pump is about 266/36000 = 0.007 or 0.7% of the collected energy.
This small efficient DC pump can easily be run off a small PV panel if the 0.7% if important is worth it to anyone.

The tank is large, and does not change in temperature that much over the course of a day, so whether you make the solar fraction calculation at the exact time of a draw or in the afternoon (when we use most of our hot water) would really make little difference. The typical drop in tank temperature from the afternoon peak to the next morning low is about 8F, and on the great majority of days, it never drops below the 110F we set as our hot water target temp, so no backup energy is used no mater when in the day a draw is made. Most of the year the backup heater is completely turned off.

Our water draws are quite predictable, and that combined with the fact that the tank temperature changes very slowly makes it extremely unlikely that a large draw would be made when the tank temperature is low -- at quick look at the temperature plots would convince you of this.

There really is no need for a flow meter if the tank temperature does not change much during a day -- just assuming that all water is drawn at a fixed time is an perfectly adequate assumption.

The aim of logging the system and estimating the solar fraction is not to get the solar fraction down to the nearest half percent -- its to show that a homeowner built system that costs 1/8 th of a typical commercial system can perform at levels equal to or better than most commercial systems do. Oversizing the collection area and storage volume while optimizing tilt angle for winter performance allow this system to perform through the winter at solar fraction levels near 90% when most commercial systems sized to the usual rules and collector tilt angles are happy to be at 50%.


Oct 4, 2009 5:35 PM ET

Very useful DIY tips
by Energy Efficient Mortgages

My husband and I have been contemplating try to build a solar hot water system. Your tips and comments are very useful. I also work with people looking to use green mortgages to finance their home purchase and solar projects. I will direct those who are interested in a solar hot water system to this article.

Jan 8, 2010 5:22 PM ET

Question about EPDM Liner used in Gary's Tank
by Anonymous

I've tried contacting you on your site.. but not heard anything back.. then i saw your postings here and thought i'd try to contact you here. I've started a Solar Tank using your Design. I have the collectors built and have bought the controller, But when i went to buy the EPDM Liner.. i can't find any listed for solar tanks. They have them listed for roof, ponds, canals, etc.. Is there a difference in these.. or is EPDM the same for all.. and I didn't see where you told what thickness you used...45ml or .60ml.. Any information you can provide would be greatly appreciated.

Mar 14, 2010 8:21 AM ET

hybrid flat plate + evacuated tube collector systems
by Anonymous

Do you have an opinion on SHW systems that use a flat plate collector in combination with an evacuated tube collector? (described in Home Power HP135, Feb-Mar 2010, pp 82-86 - "SHW Retrofit")? I'm looking for a SHW system for new construction in MA.

Jun 10, 2010 8:51 PM ET

re: Question about EPDM Liner used in Gary's Tank
by Gary Reysa

Hi -- Sorry I missed your comment, and I suppose this is too late to help.
I use the same EPDM that is used for landscape pool liners or roofing. Its sold in many places -- here in Bozeman even the Home Depot has it.
The stuff I use is 45 mil.

Don't know why I did not get your email -- I make a point of answering every email I get.


Sep 24, 2010 3:30 PM ET

Fafco Hot2o system
by Metal Man

I installed the Fafco system in July of 2008 and have been very happy with it. It's no secret it's not going to be as efficient as tubes or many other systems, it is relatively cheap and easy to install. I felt good about at least having a solar project going. With a water heater for a storage tank and other odds and ends I have about $2500 in it, less the tax rebate and it was about $1700. Here in central PA it's not real effective in the winter months. I have seen it run with the temps in the 20's, but it has it's limits to what it can do in cold. For me this is not an issue as I have a water heating coil in my coal stove.

Jul 8, 2011 10:38 AM ET

Solar & Wood Stove DHW
by heidi davis

In this blog you briefly describe your own DHW production system which involves both solar and a wood stove. We are designing our new home and would like to incorporate a similar approach to DHW. Could you point us in the right direction for additional design/installation methods? Do you also have a 'back up' tank (electric resistance or the like?) Any feedback is appreciated, Heidi

Jul 8, 2011 11:46 AM ET

Edited Jul 8, 2011 11:55 AM ET.

Response to Heidi Davis
by Martin Holladay, GBA Advisor

1. As I wrote in the blog, the best book on designing and installing a solar hot water system is Solar Hot Water Systems by Tom Lane. If you haven't bought the book yet, be sure to buy it.

2. My system includes a stainless-steel storage tank (a SuperStor tank) with an integral heat-exchange coil.

3. Installation of a coil in your wood stove can be dangerous if you don't know what you're doing, because these systems can explode. Your system will need at least 2 P&T relief valves, properly installed, and knowledgeable homeowners who don't accidentally turn any valves off at the wrong time. If you aren't sure how to install such a system, you shouldn't do it.

4. Yes, I have a backup system -- an Aquastar 125S propane-fired instantaneous water heater.

Register for a free account and join the conversation

Get a free account and join the conversation!
Become a GBA PRO!