I am doing a science fair project about dust buildup on solar-electric modules. I tested my modules and found that, on average, they lose 0.0332606% in efficiency per day due to soiling. Is there a graph showing efficiency losses due to dust in a more controlled setting?
Laura Ratliff • La Canada, California
Your letter made my day. It’s great to see a young person committed to science, interested in renewable energy, and using accuracies to that many decimal places!
We’ve touched on efficiency reductions due to dust dirt (called “soiling”) on PV modules a few times in Home Power articles—see the HP143 article by Justine Sanchez, which estimates an average of 5% loss per year due to soiling, and an HP127 article by Jeremy Taylor, which says that, in some especially dirty environments, modules can experience 25% losses per year!
If I take your per-day value and multiply it by 365 days per year, I get an annual loss of 12% (which doesn’t account for rainfall or array washing). That’s about halfway between the average and “especially dirty.” Lots of things might affect the amount of dirt on a PV array, such as living near a dirt road, factory, desert, or ocean; rainfall frequency; and array tilt, which can influence how much dirt sticks and how easily rain washes it off.
The value you obtained sounds plausible. Actually, if your scientific method is proper, and your experiment is set up right, then I would have to believe your numbers, at least for your specific location. Did you use two PV modules in your test: one that gets washed (called a “control”), and one that is allowed to collect dust?
You could extrapolate your numbers to create a sloping, but relatively straight, graph that would show that in a place that’s twice as dusty, there would be twice as much “negative” effect on power output, and vice versa.
See the “Back Page Basics” article by Erika Weliczko in HP131. The graph in the upper right column shows how PV module power (voltage × current) is influenced by the amount of sunlight hitting it (1,000 watts per square meter is considered normal, bright sun on a clear day). That means that 800 watts per square meter is 80% the light of a bright day. You could also consider that to be 20% dirty—it’s the same thing. Zero watts per square meter could be a pitch-black night, or it could be so much dirt that no light gets through at all.
Electrical current is affected in a very linear way relative to the amount of sunshine (irradiance, or measured over time, insolation). But voltage changes very little with irradiance. That means if you graphed irradiance on one axis and power (voltage times current) on the other axis, the line would be a pretty straight slope.
If you graph the data that you gathered from your experiment (at an average of 0.033% per day), it would start at 100% clean (100% irradiance) on one axis and 100% PV power on the other axis. The line would be plotted at the 0.033% per day rate for the number of days in your experiment, but could be extrapolated for many more days (the days between significant rainfall or array washing)—even though you don’t have the actual data. That continuation would be your scientific prediction.
You can compare your results to a 2006 study that measured efficiency losses between 46 large grid-connected PV systems in California and the desert Southwest. It was found that PV system efficiency declines by approximately 0.16% per day between significant rainfall events for systems located in urban, highway, or airport settings. The study concluded that annual energy loss was between 2.5% and 6% depending on the system location (see graph).
You are asking good questions and thinking about good ideas, so I hope this is interesting and useful for you. Let us know how your science fair project turns out!
Ben Root • Home Power
I got third place in physics at my school and third place in Engineering Research at the Los Angeles County Science Fair. I was in the Alternative Energy category at the California State Science Fair (see photo), and while I didn’t win anything at State, it was a great experience. I appreciated the help from your magazine, and it is really exciting that you want to publish my letter.
Laura Ratliff • via email