How do you clean your panels?

[J.P.M.]

After experimenting with a lot of different cleaning scenarios over the last 4 + years and making what I think is a professional attempt at measuring the results in terms of restoration of a clean condition and performance for each method, several times, I've got a lot of data that suggests (for my location), while soap and water and cloth scrubbing and rinsing/drying/windexing/buffing will get things clean(est), I've found that once cleaned that way, a simple hosing will restore almost as much lost performance (I'd estimate >75% or so) as any other method, including the more complicated methods just described.

Bottom line: Clean the array with dish soap and rinse with plain tap water from a hose. Then, every 4 weeks or so, if it doesn't rain, hose the panels with tap water straight from a hose from the top to the bottom of the array at a rate of ~~ 1.5 gal. per panel and let the array drip dry. Doing so may hold the fouling penalty to something like 2 -3 %.

DO SO EARLY IN THE MORNING BEFORE THE ARRAY GLAZING HAS BEEN HEATED BY THE SUN.

if interested, some observations/fill in info:

1.) Without rain, my array's performance falls off at a rate of approx. 1% per week due to dirt/dust/guano/etc., generally what I call array fouling, or simply fouling.

2.) Rain will clean the panels, including most of the guano smaller than an owl skrock. As might be expected, depending on the intensity and duration of the rain, it may restore all of the unfouled performance or a little, or, most likely, something in between. The recent 3.62" of rain on my array over about 2 days or so, accompanied by some, but not a lot of wind, took the fouling from about 9% performance penalty to close to zero, or nominally clean. Lots of rain gets most of the dirt off. Less rain not so much. More than that is hard/impossible to quantify.

3.) Morning dew, depending on how heavy that dew is, can have some cleaning effect. For several years I observed that the 1%/week fouling seemed to become asymptotic at ~ 6% or so and stay at about that level. I also noticed that a heavy dew will produce ~~ 0.01" to 0.02" of measureable precipitation in the rain gauge on my roof, produce roof dripping, array wetness (top and underside both BTW), wet windshields and some small puddles in the driveway. That, BTW, is about in line or reasonable agreement with what I calculate as condensation potential on a surface when nite sky radiant temps., dew points and wind are considered. On measuring array performance on such days, there was often a noticeable (but difficult to quantify in a hard way) improvement in array performance from any series of immediately prior and recent measurements. While this qualifies as only slightly better than anecdotal (if even that much), it would perhaps help to explain the asymptotic behavior of the increase in array fouling as f(time) I've logged and documented.

4.) I believe I cannot detect a difference in measured performance of an array that has been soap washed/brushed/rinsed with tap water and one that has been soap washed/brushed/rinsed with distilled water. That is, I can't find a measureable difference in performance by using distilled water, or probably D.I. water either, as a final rinse, with the distilled water rinse leaving no water spots. I've observed hard water spots left behind by tap water rinsing as Mike notes, but as hard and as much as I've looked and combed the data I've collected, I can't see a performance or fouling penalty from hard water spots. If it does exist, which I kind of doubt, I'm pretty comfortable saying it's probably less than the 0.75% or so that I feel comfortable claiming as my +/- tolerance for array fouling estimates. If so, that 4 week hosing schedule I suggest will probably take care of it, or at least make it hard to measure. As it turns out, the transmittance of group IIA chemicals that mostly make up the stuff that makes water hard and leave spots, is pretty high in the wavelengths of solar radiation that PV panels need. That makes the eye not the best judge of what is happening in terms of what hard water spots do to PV performance.

Take what you want of the above. Scrap the rest.

[Mike90250]

I just use a hose with a jet nozzle (sweeper nozzle) It just barely gets the stream lofted above the highest panels, and just the impact of the water gets the bird poop off. The panels are at enough of an angle to let water run off, don't need a squeegee. if you have hard water, one of the DI Rinse car wash kits to prevent spotting might be useful.

[reader2580]

I need to clean the panels on the roof of my detached garage. How do you guys clean your panels? The entire one side of the my roof (32 feet wide by 17 feet) is covered with panels. The city and fire chief allowed me to cover the whole side of the roof with no walkways since they don't vent detached garages in case of fire.

I was looking at 16 foot painter's poles at the store yesterday, but none of them seem very sturdy at 16 feet. I didn't buy one because the window cleaning heads with threads would not fit on any of the really long poles. I was thinking about standing on the other side of the roof to clean, but then I couldn't use the squeegee to get the water off. I would like to avoid cleaning from a ladder if possible.