cleaning panels

[Mike90250]

I wash my panels right at dawn, while they are still wet from dew, and before they get hot. I loft water from a garden hose and just spray them down. Weight of falling water is good enough.

Don't wash hot panels, ever see a cracked car windshield with just 1 zig zag crack - thermal stress

[littleharbor]

Wow!, large system. I wouldn't mess with it, being that large but you'd probably be surprised at how little pollen, dirt, dust affect the output. If you had two identical panels, side by side, the only difference being one was cleaned and the other dirty the output would be very close. When looking at the array from a very shallow angle, basically looking across the glass it can look pretty bad when dirty. I see the same type thing when looking out across the hood of my truck. I'm not advocating, not cleaning your panels, just saying it really has minimal affect till the fouling gets bad. Now, bird poop, that's another story, it can severely affect output if allowed to get out of hand.

[J.P.M.]

Performance will never be as good as the day of startup, and it will rarely, if ever be => STC rating.

However, on your cleaning question: PRELIMINARYSTUFF: AFTER USUAL SAFETY PRECAUTIONS, DO NOT POWER WASH ! ALSO, NEVER SPRAY WATER ON A WARM/HOT ARRAY ! CLEAN IN THE MORNING BEFORE THE SUN GETS TOO HIGH.

After a long period of no cleaning, mild soap like "Dawn", etc. and a soft cloth on a long pole of the type used to clean windows will work just fine. Then rinse with a hose. Do not use abrasive cleaners or chemicals. They will PERHAPS damage the ARC coating on the glazing. Elbow grease only. For sap/pine resin, etc, some solvent may be necessary but go slow and try soap first ,and in spite of the temptation, never use a blade. And - it goes without saying - DO NOT WALK/PUT WEIGHT ON A PANEL.

I've got a lot of data from 4+ years of a lot of measurements of performance under very clear skies and rates of fouling under various conditions and cleaning scenarios. Long, boring story. Abstract of some of what I think I've found with respect to performance and dirt on my array:

1.) Removing hard water spots w/a D.I. etc. rinse or special treatments after washing makes no measurable improvement in an array's performance over a hose/tap water rinse, at least that I've been able to observe/measure. That is, before/after cleaning incremental performance improvement is the same whether or not there are hard water spots left on an array. Skip the D.I. cost/hassle. The eye is a poor indicator of array performance as f(water spots).

2.) Array fouling rates are all different depending on location and climate. For example, the dirt on an array next to a freeway (with probably a lot of tire dust for example) is not the same as dirt in my neighborhood. Or, my array's fouling characteristics are different than the big array in Death Valley at Furnace creek, at least according to the data the folks there have shared with me (FWIW, they commercial clean their MW array every 2 years and don't worry about dust/guano) Their array fouling rates seem to be asymptotic at ~ 7-10 % of clean performance, but IMO, their monitoring is poor.

3.) I'm probably a pretty typical suburban location. FWIW, without rain, my array's performance deteriorates ~ 0.75% -1.0% per week as it fouls up. Your rate may be similar or more variable. In the absence of rain or cleaning, I believe my rate of fouling may also tend to be asymptotic as f(time), but weather has not allowed me to do much with that, and I've not seen anything in the literature that discusses the matter.

4.) After a good cleaning as described above, I've found a simple hose rinse at a rate of ~ 3/4 gal./panel with no wiping restores ~ 3/4 or so of the clean performance of my array. So, after 4 weeks of no rain, if my performance penalty from fouling is, say, 4 %, I might get the performance back up to 99 % or so of clean with a 5-10 minute or so hose rinse, or close to 100% with a trip up to the roof, and 45 minutes with water/soap/elbow grease. When I'm not measuring clean performance etc. by daily deep cleaning, I rinse w/ a hose after 4 weeks of no rain and accept an average 2-3 % fouling penalty.

5.) A moderate /decent rain, say, maybe 0.25 in or more precip. restores ~ 2/3 - 3/4 of the performance lost to fouling.

6.) Guano penalty depends on guano size. House finch drops do not seem to impair array performance. The big owl skrocks, etc. might have some measurable effects on performance, but I've not been able to measure such effects. Still, I'm very happy cows don't fly.

Bottom line: If your situation is somewhat/sort of like mine, clean the array 1X/yr. or so w/ soap and a cloth, and then rinse 1X/month or 4 weeks or so after a decent rain or since the last rinse. I'd skip commercial cleaning all together. Unless you're disabled, it's usually a rip off.

[freshtry]

Thanks for the quick response. Dual Axis Tracker would not work well in our Florida environment. Much too much wind. We just added 29 S-energy 360w panels to supplement our system. Have total of 37.1 kw at this point. Old (1 year old) panels have distinct grey appearance compared to new panels. Considering foam cannon with bio-degradable soap to clean panels, perhaps bi-annually.

[littleharbor]

Answer, Cold clear weather and a dual axis tracker.
Panels rarely put out full rated power. Warm or hot PV cells produce less voltage than cold cells.
The angle of the panels to the sun will have a direct affect on the current the panels put out. A dual axis tracker is about the only way to get maximum output all day. Nowadays solar panels have become so cheap that the more cost effective thing to do is take the money you'd spend on a dual axis tracker and buy more solar panels.
As to pollen, either dry mop the panels, if accessible, or wash them. Note, it takes quite a bit of dirt/pollen to have more than a negligible affect on power output.

[freshtry]

Spring has arrived and brought oak pollen, dust and grime to the surfaces of our solar array. It appears to be affecting the energy absorption as our system is not performing nearly as well as last year. What are the various methods/solutions to keep the panels performing close to specs?