Monday I summarized the preliminary data gathered from the solar panel prototypes (see Part 1 here). In this note, I will look at each panel and discuss its strengths, weaknesses, and performance.
In the manifold collector, water enters the bottom and is distributed into risers connected to a sheet of metal. This collector was easy and quick to build, requiring only sweat soldering and bending of sheet metal in addition to basic carpentry skills.
In the test, the 8 ft² manifold collector produced 2152 BTUs over the 9 hr test period. For each square foot of panel, it averaged 0.50 BTUs per minute with a maximum of 1.63 and a minimum of -1.447 (during a cloudy period in the afternoon). For a sense of what this felt like, water passing through this panel increased an average of 2.2°F. At its best, it raised the temperature 5.9°F and at its worst, it shed 6.3°F.
Water in this panel is distributed from a top bar and runs down over what is essentially a hot tin roof before being collected in a gutter. This is a nice idea, but the uncontrolled flow of water over the corrugated metal posed two problems: when hot, steam was produced, which condensed on the glazing of the panel, reducing its efficiency; and it was difficult to seal the gutter to the metal panel and this caused slow leaks. The overall construction was a little more complex than the manifold collector, but it did not take much more time. It involved sweat soldering, PVC pipe assembly, and basic carpentry skills.
This panel gave negative BTUs for most of the day. In sum, it averaged -0.7 BTUs/min/ft², resulting in 3115 BTUs lost over the 9-hr testing period (range: -12.5–3.4 BTUs/min/ft²). This does not mean the panel would work as a refrigerator, but just that its maximum working temperature is somewhere around 130°F. When the water entering the panel was below this threshold, it was able to raise the water temperature by a few degrees, in fact, at low temperatures it out performed the other designs, but once the water got hot enough to put out steam, the panel acted to cool the incoming +130°F water instead of heating it.
The semi-concentrated panel was set up like the manifold collector except that instead of metal panels absorbing heat and transferring it by contact with the riser pipes, it had parabolic troughs of reflective material, which concentrated the sun’s light onto the riser bars. It was more complicated to build than the manifold and trickle-down collectors because the parabolic troughs required more construction and geometry skills.
This collector did not perform as well as the manifold collector but it was much better than the trickle-down design. It averaged 0.14 BTUs/min/ft² with a high of 1.1 and low of -2.0 BTUs/min/ft². Over the 9-hr test period, it produced 609 BTUs.
This was the only panel with performance similar to the basic manifold collector. Water rises through a single pipe bent back and forth across a metal heat-collecting sheet. This was about as easy to build as the manifold collector: sweat welding, bending sheet metal, and basic carpentry.
It was able to produce an average of 0.45 BTUs/min/ft² for a total of 1924 BTUs for the test session. It had a greater range than the manifold collector, at -2.8–8.0 BTUs/min/ft².
Panel Choice and Next Steps
The clear choice for the final design is the manifold collector. In addition to putting in the best performance numbers in a side-by-side comparison, the manifold collector is better suited for a freezing-temperature location. The trickle-down and circuitous collectors might trap pockets of water instead of allowing for free drainage at night, which could cause pipes to burst. In a non-freezing environment, users will have a greater diversity of designs from which to choose, from any of the above to large tanks of water heated directly by the sun.
The next step involves optimizing the manifold collector design. We need to achieve something closer to 1.0 BTUs/min/ft² — double what the prototype achieved. The first place to look for improvement is creating a better connection between the sheet-metal absorber and the riser pipes. In the test unit, the sheet was wired to the pipes. Soldering would be ideal but as the absorber sheet is aluminum and the pipes are of copper, they expand at different rates and thus any soldered welds would fail. Instead, I’ll build a bending jig that will wrap the sheet tightly around the pipe, giving more surface area contact. Additionally, the number of risers can be increased from one every 6 in to one every 3 in. It may be necessary to choose a different glazing material, but we’ll make that decision after testing the tweaks that do not significantly raise the price of each panel.
“Lab Notes” are a series of posts chronicling the daily progress our research projects. Research Project No. 1 is the testing and installation of a solar heating system for domestic water and space heating. These notes may be useful for anyone interested in building such a system at home. Others might prefer the more succinct guide to solar heating, videos, and other formal publications that will result from this research project and be posted to the website as they are available.