Off The Shelf DIY Solar Water Heater -- Glazed

This section covers adding glazing to the Off The Shelf Domestic Solar Water Heating System.   The idea is to see if a simple, easy to make, and inexpensive glazing scheme can be developed to improve the performance of the collector in cold and part sunny weather.

off the shelf diy solar water heating collector


Leaky Glazing

This section describes a way to provide glazing for the domestic solar water heater that uses a pool heating collector.  The idea of the glazing is increase the efficiency of the pool heating collector in order to get it to perform better in low ambient temperatures and in partial sun.  This is mainly of interest to people on cold climates.

The glazing is fitted with some intentional openings that allow a limited amount of air to circulate between the glazing and the collector.  The idea is that this air circulation will keep the absorber temperature down to acceptable levels when the collector is stagnated (no water flow) during sunny periods.  The collector is made from polypropylene which has maximum service temperature of around 250F (various sources quote max service temperature from 210F to 275F).  The goal is to hit an airflow that significantly improves the collector efficiency under cold conditions but also keeps the collector surface temperature below 250F (preferably lower) during stagnation events.

The glazing is very simple.  It is SunTuf corrugated polycarbonate glazing, which is often used for applications like patio roofs -- its tough, long lived, and has a high service temperature.  The collector is 4 by 10 ft, and the glazing is made from two 12 ft by 26 inch sheets of SunTuf that are spliced together in the long direction to make a single sheet.  The length is cut to extend just beyond the collector manifolds on each end. 

The glazing is supported off the surface of the collector by two cedar 1 by 4's that rest on the roof on either side of the absorber.   The 1 by 4 lengths are such that they just fit between the collector manifolds.  

At the ends of the SunTuf sheets, and just beyond the manifolds, a sheet of 1 inch rigid polyisocyanurate insulation board closes off the end of the collector and insulates the manifolds.  The insulation board is held in place by foaming it with Great Stuff foam to the glazing.

To keep the glazing from sagging inward in the middle toward the collector and to steady the glazing in the wind, two half inch lengths of galvanized EMT electrical conduit are run across the collector just under the glazing.

The weight of the glazing assembly is supported by the lower manifold.  The glazing assembly is held down onto the collector with two straps that pass over the glazing and are screwed to the roof (similar to the way the collector itself is supported).  The attachment points of the collector straps and the glazing straps could be the same.

The entire glazing assembly for the 4 by 10 ft collector is light weight and can be managed by one person. 

solar collector glazing
The glazing panel before the rigid foam board
end panels are applied.

Showing the two edge 1 by 4's, the center
glazing splice strip, and the foamed in place
end panel of insulation board.

The insulating panels on the ends of the
collector.  Bonded to the collector with
polyurethane foam (Great Stuff).


Shows one of the two EMT conduits going
across the collector to support the glazing and
to keep it from flopping around in the wind.
The EMT is installed through holes in the 1 by 4
drilled such that the EMT is in contact with the
bottom of the SunTuf corrugations.

The glazing in place on the collector.
The orange and black polypropylene ropes
hold it down to the collector.

The glazing assembly weight is supported by the lower
manifold as shown above.
The short unpainted wood piece is a handle
to maneuver the glazing into place -- can be
removed once the glazing is positioned.
The spaces around the manifold allow some
air to flow under the glazing for
stagnation ventilation.

The short section of SunTuf added to the bottom
of the glazing is used because an earlier
version of the glazing was shorter, and I did
not want to buy new longer pieces of SunTuf -- on
a final version of the collector, the glazing
would run full length.

The air velocity through the openings near each manifold was about 200 fpm -- in at the bottom and out at the top.  If the area is about 4 sq inches, then the total flow into the bottom of the collector is about (8/144)(200) = 11 cfm.  If this air comes in at 40F and leaves at 110F, then its carrying away about (11 cf/min)(60 min/hr)(110F-40F)(0.061 lb/cf)(0.024 BTU/lb-F) = 670 BTU/hr  -- of the order of 1/10th of the collector heat output per hour.

This glazing scheme is simple and easy to build.  It was designed mostly just to test whether having the glazing significantly improves the collector performance.  One problem that I see is that the insulating boards that close off the ends might be subject to damage is displacement by snow.  There are probably glazing designs that would look and perform better -- any ideas?


The plot below shows one sample day of testing with the glazing described above. 

performance of off the shelf solar dhw system

In this plot:

- Red is collector supply temperature (F) -- from the tank

- Solid Blue is collector return temperature (F) -- back to tank

- Solid Green/Yellow is solar intensity (w/sm)

- Dash Purple is collector surface temperature

- Dash Green is ambient temperature.

If you look at the sun curve, it may seem strange that there are sun levels that go above 1200 watt/sm, in that 1000 watts/sm is often quoted for full sun.  The main reason for this is that the pyranometer is mounted in the plane of the glazing (so that it will receive the same radiation the the collector receives), and in this position it gets both direct sun and sun reflected of the snow field in front of the collector.  Pyranometers at weather stations are mounted horizontally, and do not get the snow or ground reflections.

The points that define the efficiency curve for the for the glazed panel in the plot below are taken from several days of testing similar to the one above.  The points to estimate efficiency were taken at points where the collector was experiencing steady conditions. 

collector efficiency comparison

The plot above is a comparison of the efficiency of:

  1. Red Squares --  unglazed pool heating collector,
  2. Blue Diamonds -- a high quality, copper fin and riser, commercial glazed flat plate
  3. Green Triangles --  our glazed pool heating collector with some air leakage to reduce stagnation temperatures.
  4. Purple X's -- Evacuated tube collector

The horizontal axis parameter is the temperature difference between the collector absorber and the ambient temperature divided by the solar intensity.  Basically, as it gets colder and/or the sun gets less intense, the efficiency of all these collectors drop -- but some drop faster than others.

The curves for the unglazed collector, commercial flat plate collector, and evac tube collector are taken from the SRCC certification data.  The curve for our glazed pool collector is developed from my testing.  So, given that I don't know the details of the SRCC test procedure, there may be some differences just due to differences in methods.

As expected, over most of the range of sun and temperatures of interest for water heating, the glazed pool heating collector is better than the pool heating collector, but not as good as the commercial flat plate.  The reasons its not as good as the flat plate could be:  1) the intentional air leakage under the glazing reduces its efficiency, 2) the absorber construction of the pool heating collector is probably not be as efficient as the copper based collector, 3) differences in the procedure for determining efficiency between the SRCC methods and my testing.

The table below shows the same data related to common operating conditions for solar water heating.  The points in the table assume that the solar water tank temperature is 105F and the average absorber temperature is 110F. 

Weather Ambient Sun Parm Unglazed Pool Our Glazed Pool Glazed Flat Plate Evac Tube
Condition Temp (F) BTU/sf-h Tt-Tc/Isun BTU/sf BTU/$ BTU/sf BTU/$ BTU/sf BTU/$ BTU/sf BTU/$
70F -- Full Sun 70 300 0.133 131.9 38.5 142.9 26.3 175.0 5.6 146.7 4.8
50F -- Full Sun 50 300 0.200 65.6 19.1 102.8 18.9 153.6 4.9 140.8 4.6
30F -- Full Sun 30 300 0.267 0.0 0.0 62.6 11.5 132.2 4.3 134.8 4.4
70F -- Part Sun 70 210 0.190 52.5 15.3 75.9 14.0 109.7 3.5 99.1 3.3
50F -- Part Sun 50 210 0.286 0.0 0.0 35.8 6.6 88.3 2.8 93.2 3.1
30F -- Part Sun 30 210 0.381 0.0 0.0 0.0 0.0 66.8 2.2 87.2 2.9

I've colored the text green for the winning collector on heat output per sqft of panel.

I've colored the text purple for the winning collector on BTU pers sqft per dollar of collector cost -- probably the thing you are most interested in.

The BTU produced per dollar of collector cost are an indicator of the cost effectiveness of the collector -- the more BTU per dollar, the better.

Every type of collector wins at something!

Some messages from this:

Here is yet another way to look at the same data:

It would be nice to do an actual simulation of yearly energy dollar saved for all the collectors in a few climates.  Anyone want to take this on?-- would make a great high school or university project.

It seems to me that for most of the conditions on the chart, the glazed pool heating collector beats the two commercial collectors on heat output per dollar of collector cost by a factor 3 to 5.

An observation:   A lot of people think that the climate they live in is a lot colder than it really is -- don't trust your gut on this, go to a site like and actually look at what your average daily high temperatures are.  I am in cold Montana that once is a while gets below -30F at night, but we no months where the average daily high temperature is not above freezing.  Remember, you do the most solar collecting near the daily high temperature.

Sealed Glazing

I sealed up the openings that were allowing some air to flow under the glazing with duct tape.  Very ugly, but I think that it is pretty effective from a selaing point of view.

glazing vents sealed up


glazed collector efficiency

This is the same plot as above with some points added for the one day of sealed glazing tests.

Purple Triangles -- The sealed glazing

All the rest of the data is the same as the plot above.

So, generally the sealed glazing points lie on the good (more efficient) side of the the line for the glazed (but not sealed) collector.  The gain for sealing the glazing averages about a 6% increase in efficiency, which is good for about a 20% increase in heat output for this area of the efficiency curve. 

A 20% gain does seem worthwhile, but the downside is that some form of automated opening vent would likely be required to keep the collector from being damaged in a stagnation event.

I'm going to leave the collector sealed and see what the collector surface temperatures run on warmer days.  

Update: March 21, 2013: We had a part sunny day today with a gusty wind at about 10 mph -- ambient temperature about 30F. I noticed that the sealed collector was running and making some hot water. I think that in these windy conditions, the unsealed and certainly the unglazed collector would not have gotten warm enough to turn the pump on. So, better wind protection for the collector with the sealing is something to think about in places where some wind is common.

I'd appreciate any thoughts, ideas, comments, questions -- please use this page for comments...


Gary  March 16, 2013